COMPUTER

NETWORKS

Nastaran Nazar Zadeh

TAP
Toronto Academic Press
COMPUTER NETWORKS

Nastaran Nazar Zadeh

Toronto Academic Press

4164 Lakeshore Road
Burlington ON L7L 1A4
Canada
www.tap-books.com
Email: [email protected]

© 2025
ISBN: 978-1-77956-708-6 (e-book)

This book contains information obtained from highly regarded resources. Reprinted material
sources are indicated and copyright remains with the original owners. Copyright for images
and other graphics remains with the original owners as indicated. A Wide variety of references
are listed. Reasonable efforts have been made to publish reliable data. Authors or Editors or
Publishers are not responsible for the accuracy of the information in the published chapters
or consequences of their use. The publisher assumes no responsibility for any damage or
grievance to the persons or property arising out of the use of any materials, instructions,
methods or thoughts in the book. The authors or editors and the publisher have attempted
to trace the copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission has not been obtained. If any copyright holder has not been
acknowledged, please write to us so we may rectify.

Notice: Registered trademark of products or corporate names are used only for explanation and
identification without intent of infringement.

© 2025 Toronto Academic Press

ISBN: 978-1-77956-291-3

Toronto Academic Press publishes wide variety of books and eBooks. For more information
about Toronto Academic Press and its products, visit our website at www.tap-books.com.
ABOUT THE AUTHOR
Nastaran Nazar Zadeh is a highly experienced computer engineer,
researcher, and advisor in the fields of robotics, artificial intelligence and
computer science. She holds a Master of Science in Computer Engineering
from Mapua University of the Philippines and pursed her Ph.D. in
Electronic Engineering at the same institution. With over seven years
of teaching experience, Nastaran has taught electronic and computer
engineering programs at several reputable academic institutions,
where she has also led numerous thesis studies. Her research focuses
on developing robotics systems with A.I. and machine learning, which
enables her to stay up-to-date with the latest advancements in the field
and implement cutting-edge technologies.
 Table of Contents

List of Figures ix
List of Tables xiii
Preface xv

1 COMPUTER Review Questions 48

References 48
NETWORKS
AND SERVICES 1
2 NETWORK
Introduction 4 MODEL AND
1.1. Fundamentals of Computer Networking 5
STANDARDIZATION 49
1.1.1. Open Systems Interconnection (OSI) 6
1.1.2. Protocol 7 Introduction 51
1.1.3. Network Topology 9 2.1. The OSI Model 52
1.1.4. Unique Identifiers of Network 14 2.1.1. Application Layer 53
1.2. Types of Network Services 18 2.1.2. Presentation Layer 53
1.2.1. User Management 19 2.1.3. Session Layer 54
1.2.2. Email Management 22 2.1.4. Transport Layer 55
1.2.3. Printer Sharing 24 2.1.5. Network Layer 56
1.2.4. System Administration 26 2.1.6. Data Link Layer 57
1.3. Types of Network 28 2.1.7. Physical Layer 58
1.3.1. Local Area Networks (LAN) 29 2.2. Multiplexing and Switching 60
1.3.2. Personal Area Networks (PAN) 30 2.2.1. Types of Multiplexing 61
1.3.3. Home Area Networks (HAN) 31 2.2.2. Switching 63
1.3.4. Wide Area Networks (WAN) 32 2.2.3. Digital Multiplexers 64
1.3.5. Campus Networks 32 2.2.4. Digital Demultiplexers 66
1.3.6. Metropolitan Area Networks (MAN) 33 2.2.5. Multiplexers as PLDs 67
1.3.7. Enterprise Private Networks 34 2.3. Transmission Control Protocol (TCP) 69
1.3.8. Internetworks 37 2.3.1. TCP Header Format 71
1.3.9. Backbone Networks (BBN) 38 2.3.2. Connection Establishment and
1.3.10. Global Area Networks (GAN) 39 Termination 73
Summary 46 2.3.3. Sliding Window and Flow Control 75
Multiple Choice Questions 46 2.3.4. Congestion Control 75
 2.4. User Datagram Protocol (UDP) 78 Summary 136
2.4.1. Protocol Header 79 Multiple Choice Questions 136
2.4.2. TCP over Wireless Links 80 Review Questions 137
2.4.3. Characteristics of Wireless Media 80 References 137
2.4.4. TCP Performance Enhancements 81
2.4.5. Wireless System Evolution and TCP 82
4 INTERNETWORKS 139
2.4.6. IP Networks 83 4.1. Principles of Internetworking 141
2.4.7. Network Security and Performance 4.1.1. Principle 141
Analysis 85
4.1.2. Challenges in Internetworking 143
Summary 94
4.1.3. Internetwork Addressing 143
Multiple Choice Questions 94
4.1.4. Unit of Internetworking 147
Review Questions 95
4.2. Internetwork Architecture 150
References 95
4.2.1. Application Layer 154
4.2.2 Transport Layer 156
4.2.3. Internet Layer 159
3 LOCAL AREA
4.2.4. Network Access Layer 160
NETWORK/WIDE 4.3. IP Addressing and Architecture 166
AREA NETWORK 97 4.3.1. Version of IP address 167
4.3.2. IP Address Types 167
Introduction 99
4.3.3. Types of Website IP address 168
3.1. Local Area Network 100
4.3.4. IP Address Classification Based on
3.1.1. The Evolution of LAN 101 Operational Characteristics 168
3.1.2. LAN Technology 101 4.3.5. IP Addresses Work 169
3.1.3. LAN Applications 102 4.3.6. Static IP Address 171
3.1.4. Characteristics of a LAN 103 4.3.7. Track and Record IP Addresses 172
3.1.5. Devices 105 4.3.8. Difference between Public and
3.1.6. Hardware Devices for LAN 106 Private IP Addresses 173
3.1.7. LAN Advantages and Services 107 Summary 178
3.1.8. Protocol 108 Multiple Choice Questions 178
3.1.9. LAN Standards 109 Review Questions 179
3.2. Wide Area Network 112 References 179
3.2.1. WAN Protocol 113
5 NETWORKING
3.2.2. WAN Devices 117
3.2.3. Connecting Your Wide Area
FOR CLUSTERS 181
Network (WAN) Devices 118
Introduction 183
3.2.4. WAN Management and
Optimization 121 5.1. Cluster-Based Networks 184
3.2.5. Types of WAN Connections and 5.1.1. Cluster Based Network
how WAN Connections Work 122 Architectures 185
3.2.6. Wide Area Network (WAN) 5.1.2. Routing in Cluster-Based Networks 191
Standard 123 5.2. Clustering Protocols 198
3.2.7. WAN Port 123 5.2.1. Classification of Clustering
3.2.8. Build a WAN 124 Protocols 198
5.2.2. LEACH Clustering Protocol 199
 5.2.3. Clustering and Routing Protocol 6.4. Characteristics of a Distributed System 244
for Wireless Sensor Networks 200 6.4.1. Fault-Tolerant 245
5.2.4. Prepare Cluster Networking 6.4.2. Scalable 245
Infrastructure 205
6.4.3. Predictable Performance 246
5.3. Networks and Inter-Connection/
Switching Devices 209 6.4.4. Openness 247

5.3.1. Network Interface Card (NIC) 209 6.4.5. Security 247

5.3.2. Repeater 210 6.4.6. Transparency 247

5.3.3. Hub 210 Summary 253

5.3.4. Bridge 210 Multiple Choice Questions 253

5.3.5. Switch 215 Review Questions 254

5.3.6. Router 216 References 254

Summary 218 7 WIRELESS

Multiple Choice Questions 218
NETWORK SECURITY 257
Review Questions 219
References 219 Introduction 259

6 DISTRIBUTED 7.1. Synopsis of

Wireless Technology 260
NETWORK SYSTEMS 221 7.1.1. Wireless Networks 260
7.1.2. Wireless Devices 263
Introduction 223
7.1.3. Wireless Security Threats and Risk
6.1. The Client-Server Model in a Mitigation 265
Distributed Computing System 224
7.1.4. Emerging Wireless Technologies 267
6.1.1. Features and Problems of the
Client-Server Model 225 7.1.5. Federal Information Processing
Standards 268
6.1.2. Cooperation between Clients
and Servers 226 7.2. Development of Wireless LANs 269

6.1.3. Cooperation Type and Chained 7.2.1. Purpose of Wireless LAN 270
Server 227 7.2.2. Benefits 273
6.1.4. Multiple Servers 227 7.2.3. Security of 802.11 Wireless LANs 274
6.2. Extensions to the Client-Server Model 230 7.2.4. Security Requirements and Threats 284
6.2.1. Agents and Indirect Client-Server 7.2.5. Emerging Security Standards and
Cooperation 231 Technologies 288
6.2.2. The Three-Tier Client-Server 7.3. Wireless Personal Area Networks 290
Architecture 233 7.3.1. Bluetooth Technology 290
6.3. Service Discovery 237 7.3.2. Benefits 294
6.3.1. Hardwiring Computer Address 238 7.3.3. Security Requirements and Threats 296
6.3.2. Broadcast Approach 238 Summary 302
6.3.3. Name Server Approach 239 Multiple Choice Questions 302
6.3.4. Broker-Based Location Lookup 239 Review Questions 303
6.3.5. Client-Server Interoperability 240 References 303
6.3.6. Client/Server Computing 241
INDEX 305
List of Figures

Figure 1.1. Example of computer network parts

Figure 1.2. Format of computer network topology
Figure 1.3. Functionality of bus topology
Figure 1.4. Structure of ring topology
Figure 1.5. Arrangement of star topology
Figure 1.6. Networking of mesh topology
Figure 1.7. Features of mesh topology
Figure 1.8. Structures of hybrid topology
Figure 1.9. Unique identifiers of network
Figure 1.10. Identification of active network connection
Figure 1.11. Translation of web addresses or URLs (ex: www.google.com) into corresponding
IP addresses
Figure 1.12. Administrators to manage users and the actions they are permitted to take on
objects
Figure 2.1. The request context of application layer
Figure 2.2. Presentation layer
Figure 2.3. Session layer
Figure 2.4. Transport layer in OSI networking system
Figure 2.5. Network layer process
Figure 2.6. Framing of data link layer
Figure 2.7. Physical layer
Figure 2.8. Multiplexing
Figure 2.9. 4-to-1 multiplexer
Figure 2.10. Digital demultiplexers
Figure 2.11. TCP header format
Figure 2.12. TCP connection establishment
Figure 2.13. TCP connection termination
Figure 3.1. Example of a typical LAN application
Figure 3.2. Throughput and rate of transmission of the frame
Figure 4.1. Each interface on a device is uniquely identified by a data-link address
Figure 4.2. MAC addresses, data-link addresses, and the IEEE sub-layers of the data-link layer
are all related
Figure 4.3. The MAC address contains a unique format of hexadecimal digits
Figure 4.4. Each network interface must be assigned a network address for each protocol
supported
Figure 4.5. Internetworking flow meter
Figure 4.6. How information travels on internet?
Figure 4.7. UML class diagram of transport layer
Figure 5.1. The link-clustered architecture
Figure 5.2. The NTDR network architecture
Figure 5.3. The virtual subnet architecture
Figure 5.4. Direct routing
Figure 5.5. Long-path routing
Figure 5.6. The nested cluster architecture
Figure 5.7. Quasi-hierarchical routing versus stricthierarchical routing
Figure 5.8. The link-state routing architecture with gates
Figure 5.9. The regional node routing architecture
Figure 5.10. The SURAN routing architecture
Figure 5.11. The MMWN routing architecture
Figure 5.12. Basic network setup
Figure 5.13. Add 3rd NIC of each cluster member to existing VLAN ‘CICN’
Figure 5.14. Create a new dedicated and isolated VLAN
Figure 5.15. Bridge configuration message format
Figure 5.16. A simple source-route bridging network
Figure 6.1. The basic client-server model
Figure 6.2. Cooperation functionality between client and server
Figure 6.3. Printing service (a service example)
Figure 6.4. Indirect client-server cooperation
Figure 6.5. Examples of three-tier configurations
Figure 6.6. An example implementation of the three-tier architecture
Figure 6.7. Service discovery -- broadcast approach

x
Figure 6.8. Transparency in distributed systems
Figure 7.1. Independent and infrastructure basic service sets
Figure 7.2. Extended service set
Figure 7.3. WEP encryption/decryption process. (black) common steps, (Red) encryption
process, (Blue) decryption process
Figure 7.4. Taxonomy of 802.11 authentication techniques
Figure 7.5. Shared-key authentication message flow
Figure 7.6. Taxonomy of security attacks
Figure 7.7. Bluetooth operating range
Figure 7.8. Man-in-the-middle attack scenarios

xi
List of Tables

Table 7.1. Key characteristics of 802.11 wireless LANs

Table 7.2. Key problems with existing 802.11 wireless LAN security
Table 7.3. Key characteristics of Bluetooth technology
Table 7.4. Device classes of power management
 PREFACE

In the 21st century, we have witnessed some remarkable new technologies that have significantly
altered our way of life. This significant transformation began in the late 20th century with the
emergence of the internet in 1995. The internet has introduced new methods of operation,
tools, and devices that are highly beneficial in educational institutions, businesses, and virtually
every other sector. Despite originating in 1989, the internet now hosts an immense amount of
data, approximately 1.2 million terabytes, contributed by major corporations such as Google,
Amazon, Microsoft, and Facebook. The internet is home to a vast array of content – with over 4.5
billion accessible websites, and even more content residing in the deep web, which surpasses
the size of the commonly used internet.

Shortly after the inception of the internet, email emerged, followed by a variety of other online
platforms including e-commerce websites, social media, online enterprises, digital learning,
e-government services, cloud storage, and more. These platforms began to surface between
1995 and the early 2000s. Presently, there is a multitude of internet-based technologies utilized
across various domains such as business, science, engineering, healthcare, and beyond. These
technologies have become integral to our daily lives, with most individuals utilizing them
whether by choice or necessity. Due to advancements in computing power and the decreasing
cost of data storage, both governmental bodies and private enterprises are amassing vast
quantities of information. Computer networking is about connecting computers together so
they can talk to each other and share data. Basically, a computer network is just two or more
computers hooked up together. This makes it easier for people to communicate with each other.

The aim of this book, “Computer Networks,” is to provide a comprehensive foundation in the
principles and practices of computer networking. It is designed to cater to undergraduate
students in computer science, information technology, and related fields who are seeking to
gain a thorough understanding of network concepts, architectures, protocols, and technologies.
The textbook aims to equip readers with the knowledge and skills necessary to understand the
complexities of modern networks, including the Internet, LANs, WANs, and wireless networks.

Through a blend of theoretical concepts and practical applications, the textbook introduces
fundamental networking concepts such as the OSI and TCP/IP models, network topologies,
routing algorithms, network security, and network management. It emphasizes not only the
technical aspects of networking but also explores critical issues like network security, data
communication protocols, and the latest networking trends and technologies. Intended to serve
both as a learning resource and a reference, this textbook includes numerous examples, case
studies, and exercises to reinforce learning and encourage practical application of concepts.

–Author
 CHAPTER
Computer
1 Networks and
Services

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Define the fundamentals of computer networking

• Evaluate the various types of network services
• Discuss on numerous types of network

INTRODUCTORY EXAMPLE

FANTSUAM FOUNDATION’S COMMUNITY WIRELESS NETWORK

Kafanchan is a community of 83,000 people located 200 km northeast of Abuja, in central
Nigeria. Kafanchan used to be known as a busy and thriving town as it was the host of
one of the main junctions of the national railway. When the railway industry was booming,
almost 80% of Kafanchan’s population relied on it in one way or another. Following the
complete breakdown of the Nigerian railway system, the population of Kafanchan has been
forced to go back to its original source of income, which is agriculture.
2 COMPUTER NETWORKS

Kafanchan is a poorly connected area in terms of fixed telephony and Internet

connectivity. Today, no fixed telephony (PSTN) is available in the area and GSM only
just arrived in 2005. However, the GSM coverage is just as poor as the quality of the
service. At the moment, SMS services are the most reliable communication service
because voice conversations tend to cut off in the middle and suffer heavy noise.
Poor access to electricity brings further challenges to the people of Kafanchan. The
national electric power company of Nigeria, generally known as NEPA (National Electric
Power Authority), is more commonly known to Nigerians as “Never Expect Power
Always”. In 2005, NEPA changed its name to Power Holding Company of Nigeria (PHCN).
Kafanchan is receiving power from NEPA on an average of 3 hours per day. For the
remaining 21 hours, the population relies on expensive diesel generators or kerosene for
illumination and cooking. When NEPA is available on the grid, it provides an unregulated
voltage in the range of 100–120 V in a system designed for 240 V. This voltage must
be regulated to 240 V before most loads can be connected. Only light bulbs can be
fed straight to the grid power since they can handle the low voltage without damage.

Project Participants
Given the challenging background of Kafanchan, how could anyone come up with the
idea of establishing the first rural Wireless ISP in Nigeria there? Fantsuam Foundation
did, and they made it happen. Fantsuam Foundation is a local, non-governmental
organization that has been working together with the community of Kafanchan since
1996 to fight poverty and disadvantage through integrated development programs.
Fantsuam’s focus lies on microfinance, ICT services, and social development in rural
communities of Nigeria. Becoming the first rural wireless ISP in Nigeria was part of
their mission to be a recognized leader in the provision of rural development initiatives,
as well as the foremost rural knowledge economy driver in Nigeria.
The Wireless ISP of Fantsuam Foundation, also known as Zittnet, is funded by IDRC,
the International Development Research Centre of Canada. IT +46, a Swedish-based
consultancy company focusing on ICTs for development, has worked together with
the Zittnet team to provide technical support for wireless communications, bandwidth
management, solar energy, power backup systems, and VoIP deployments.

Network Operating Center (NOC)

A new Network Operating Center was established to host the power backup system
and server room facilities. The NOC was designed to provide a place safe from dust,
with good cooling capabilities for the batteries and the inverters. The NOC uses natural
methods and is made from locally available materials. The building comprises four rooms:
a battery storage room, a server room, a working space, and a room for equipment
storage. The battery storage room hosts seventy 200 Ah deep cycle batteries, as well
as five inverters (one of them pure sine wave), two solar regulators, power stabilizers,
and DC and AC disconnects. The batteries are stacked vertically on a metal shelf
structure for better cooling.
 COMPUTER NETWORKS AND SERVICES 3
The server space accommodates a rack unit for servers and a fan. The room has
no regular windows, to avoid dust and overheating. The server room and battery room
face south to improve natural cooling and to help keep the room at an appropriate
temperature. The server room and the battery space require effective low-cost/low-energy
cooling as they need to operate 24x7. To achieve this goal, natural cooling techniques
have been introduced in the NOC design: small fans and extractors and thick walls of
bricks (double width) in the direction of the sunset. The south side of the building hosts
24 solar panels in a shadow-free area on its metal roof. The roof was designed with an
inclination of 20 degrees to host the panels and limit corrosion and dust. Extra efforts
have been made to keep the panels easily reachable for cleaning and maintenance.
The roof has also been strengthened in order to carry the extra load of 150–200 kg.
The NOC building is constructed of locally produced laterite mud bricks. The material
is cheap since it is frequently used and comes from the top layer of soil. The bricks
are produced locally by hand using a low-tech pressing technique. The NOC is unique
for its kind in Kaduna State.

Physical Infrastructure: A Communication Mast

Most potential clients of Zittnet are located between 1 km and 10 km from the premises
of Fantsuam. To reach these clients, Fantsuam established a communication mast on
their premises. In October 2006, a 45m (150-foot) tall self-standing mast was installed at
Fantsuam Foundation. The mast was equipped with grounding and lightning protection,
as well as a mandatory signal light. A metal ring was buried at the base of the tower
at a depth of 4 feet. All three legs of the mast were then connected to the grounding
circuit. A lightning rod was mounted at the highest point of the mast to protect the
equipment against lightning strikes. The rod is made of pure copper and is connected
to the earth ring at the base of the mast using copper tape. The signal light mounted
at the top of the mast is a requirement from the Civil Aviation Authorities. The light
is equipped with a photocell that enables automated switching based on the level of
ambient light. This way, the light comes on at night and goes off during the day.

Clients
The clients are free to use the Internet access for any purpose. For example, Isaiah
Balat is reselling vouchers (that he bought from Fantsuam) to his clients. His Internet
café hosts 10 computers that are all connected to Zittnet. The clients purchase vouchers
from the owner with a margin of 25% over the price offered by Fantsuam. In return,
clients that do not have access to a computer connected to Zittnet can access the
network through the PCs at Isaiah Balat’s café. The New World Hotel is another client
that aims to create a similar business model but on a larger scale. They will provide
wireless Internet access to all of their rooms and offer access to Zittnet’s uplink by
reselling vouchers. Other clients, like the General Hospital and the Jagindi Street Clinic,
are using the Internet access for professional and private use without reselling access
to their clients.
4 COMPUTER NETWORKS

INTRODUCTION
A group of linked computers is known as a computer network. Nodes refer to computers
connected to a network. Computers can be linked wirelessly using radio waves or
through cabling, typically an Ethernet cable.. Computers that are connected can share
resources such as file servers, printers, Internet access, and others. A network is a
multifunctional link that enhances the capabilities of a single computer.
A computer network is made up of two or more computers connected to one another in
order to exchange data electronically. In addition to physically joining computers and other
communication devices, a network system performs the crucial job of creating a unified
architecture that enables a range of equipment types to transfer data almost seamlessly.
IBM’s Systems Network Architecture (SNA) and ISO Open Systems Interconnection (OSI)
are two well-known architectures.
Wide-area networks (WANs) and local-area networks (LANs) are the two fundamental
types of networks. LANs use links (wires, Ethernet cables, fiber optics, Wi-Fi) that send
data quickly to link computers and peripheral devices in a constrained physical area, like
a lab, business office, or college campus. A local area network (LAN) usually comprises
multiple desktop computers, printers, and large-capacity disk storage units known as file
servers. These devices allow all computers connected to the network to access a shared
collection of files. LAN operating system software enables users to share printers and
storage devices, communicate with one another, and access centrally located processors,
data, or programs (instruction sets) all at once. It does this by interpreting input and
providing instructions to networked devices. LAN users can connect to WANs and
other LANs. Bridges, or similar devices, serve as transfer points connecting LANs with
comparable architectures. “Gates,” which change data as it travels between systems,
connect LANs with disparate architectural styles.
Wide Area Networks (WANs) link computers and local networks to larger networks
across extensive geographical regions, such as continents. Although their users typically
access the networks via a modem (a device that allows computers to communicate over
telephone lines), they may link the computers via cables, optical fibers, or satellites.
The Internet, a vast network of gateways and networks connecting billions of computer
users worldwide, is the biggest WAN.
 COMPUTER NETWORKS AND SERVICES 5

1.1. FUNDAMENTALS OF
COMPUTER NETWORKING
→Learning Objectives
• Discuss on open systems interconnection (OSI)
• Access the useful protocol for networking
• Elaborate the various network topology
• Learn about the unique identifiers of network

A computer network is an assembly of There are many types of networks,

computing hardware, including computer including:
systems, that is connected via communication
channels to enable resource sharing and • Local Area Networks (LAN).
communication among various users. • Global Area Networks (GAN).
• Personal Area Networks (PAN).
It is this network that evolved to become
what we now call the internet. • Home Area Networks (HAN).
• Wide Area Networks (WAN).
Networks are used to:
• Campus Networks.
• Facilitate communication via • Metropolitan Area Networks (MAN).
email, video conferencing, instant
messaging, etc. • Enterprise Private Networks.
• Enable multiple users to share • Internetworks.
a single hardware device like a • Backbone Networks (BBN).
printer or scanner. • Global Area Networks (GAN).
• Enable file sharing across the • The internet.
network.
It is the connection of several devices
• Allow for the sharing of software together, usually referred to as hosts
or operating programs on remote connected via multiple paths in order to
systems. send and receive media or data. Additionally,
• Make information easier to access there are various devices or mediums that
and maintain among network users. facilitate communication between two
distinct devices, collectively referred to as
network devices. Examples include Hub,
Bridge, Switch, Router.
6 COMPUTER NETWORKS

Figure 1.1. Example of computer network parts (Source: Medium.com,

https://miro.medium.com/v2/resize:fit:640/format:webp/0*rkjdV8aVrpovzJv.
png).

Network topology refers to the design pattern that links devices

together. like the Daisy chain, Ring, Mesh, Star, and Bus.

Figure 1.2. Format of computer network topology (Source:

Medium.com, https://miro.medium.com/v2/resize:fit:640/
format:webp/0*DZGU9Yd2VSSBfeSW.png).

1.1.1. Open Systems Interconnection (OSI)

It is a reference model that details the functions of every layer
Keyword as well as standards for communication protocols. A conceptual
framework for comprehending relationships is called a reference
Network device model. The OSI reference model serves as a guide for software
is hardware or
software that developers and vendors to ensure interoperability of their digital
establishes communication products and programs. It also provides a clear
and maintains framework for describing the various functions of a networking or
a secure and telecommunication system.
effective internet
connection. When describing their goods and services, the majority of telecom
vendors try to align them with the OSI model. Furthermore, although
OSI is helpful in directing discussion and assessment, it is rarely
used in practice. This is due to the fact that, unlike the OSI model,
few network products or common tools maintain related functions
 COMPUTER NETWORKS AND SERVICES 7
in clearly defined layers. The internet is to transfer data between electronic devices
defined by the TCP/IP protocol suite, which like computers. A preexisting agreement
does not map exactly to the OSI model. defining the information’s format and the
sending and receiving protocols is required
To simulate or track the transmission for computers to exchange data. Without a
or reception of data over a network, IT protocol, a computer transmitting data might
specialists employ OSI. According to this send it in 8-bit packets, whereas a computer
model, sending and receiving data is divided receiving the data might anticipate receiving
into seven layers, each of which is in charge it in 16-bit packets. International or industry-
of carrying out particular functions. wide organizations create protocols. A set
The fundamental idea behind OSI is that of guidelines for implementing networking
communication between two endpoints in a communications between computers, OSI
network can be separated into seven different (Open Systems Interconnection) is arguably
groups, or layers, of related functions. Every the most significant computer protocol.
user or program that communicates uses The norms and guidelines for
a device capable of providing those seven communication between network devices
levels of functionality. Each layer in this are specified by a network protocol.
architecture provides support to the layer Network protocols include formatting rules
above it, which in turn provides support that define the way data is formatted into
to the layer below it. Data will therefore sent and received messages, as well as
flow down through the layers of the source mechanisms for devices to recognize and
computer, across the network, and then up establish connections with one another. In
through the layers of the receiving computer addition, certain protocols facilitate data
in any given message between users. The compression and message acknowledgment
only layer in the stack that isn’t able to for dependable and/or high-performance
provide services to a higher level is the network communication.
application layer.
All contemporary computer networking
Applications, operating systems, protocols send and receive messages in
network card device drivers, and networking the form of packets, which are messages
hardware work together to provide the divided into parts that are gathered and
seven layers of function, which let a system put back together at their destination.
send a signal over a network Ethernet These packets are sent and received using
or fiber optic cable, or via Wi-Fi or other packet switching techniques. Thousands of
wireless protocols. distinct computer network protocols have
been created, each with a specific function
1.1.2. Protocol and setting in mind.

A protocol is a collection of guidelines or

formulas that specify how two entities can 1.1.2.1. Internet Protocols
communicate over a network; there are A group of related network protocols that
distinct protocols defined at every OSI model are also some of the most popular are part of
tier. DHCP, FTP, UDP, ARP, TCP, and IP are the Internet Protocol (IP) family. Higher-level
a few examples of these protocols. In the protocols like TCP, UDP, HTTP, and FTP all
field of computer science, protocol refers to interface with IP to offer extra features in
a collection of guidelines or practices used
8 COMPUTER NETWORKS

addition to Internet Protocol itself. Likewise, IP coexists with lower-

level Internet Protocols such as ARP and ICMP. Lower-level protocols
in the IP family typically communicate with network adapters and
other computer hardware, whereas higher-level protocols in the IP
family typically interact more directly with applications like web
browsers.
Remember
Computer 1.1.2.2. Wireless Network Protocols
networking is the
branch of computer Thanks to Wi-Fi, Bluetooth, and LTE, wireless networks have become
science that deals commonplace. Network protocols designed for use on wireless
with the ideation, networks must support roaming mobile devices and deal with issues
architecture,
creation,
such as variable data rates and network security.
maintenance,
and security
of computer 1.1.2.3. Network Routing Protocols
networks. It is
a combination Routing protocols are specialized protocols created especially to
of computer be used by internet network routers. A routing protocol has the
science, computer ability to recognize other routers, control the routes—the paths that
engineering, and connect network message sources and destinations—and perform
telecommunication.
dynamic routing. Routing protocols commonly used are BGP, OSPF,
and EIGRP.

1.1.2.4. How Network Protocols Are Implemented

Some network protocols are supported by software services that are
integrated into modern operating systems. Programs that support
the high-level protocols required for an application to operate, such
as web browsers, are contained in software libraries. In direct
hardware (silicon chipsets), support is implemented for a few lower-
level TCP/IP and routing protocols for better performance.
Binary data, which is made up of ones and zeros that encode
messages, is contained in every packet that is sent over and received
over a network. The majority of protocols include a brief header
at the start of every packet to hold the sender and destination
of the message. Some protocols conclude with a footer as well.
Every network protocol has the ability to recognize messages of a
particular type and handle headers and footers when transferring
data between devices. A protocol family is a collective term for a
set of interoperable network protocols at both higher and lower
levels. Traditionally, network students study the OSI model, which
divides network protocol families into discrete layers conceptually
for educational purposes.
 COMPUTER NETWORKS AND SERVICES 9
1.1.3. Network Topology
The arrangement of a network which comprises of nodes and connecting lines via sender
and receiver is referred as network topology. The various network topologies are:

1.1.3.1. BUS Topology

The bus topology is multipoint. A single, lengthy cable acts as the network’s backbone,
linking all devices. Taps and drop lines are utilized to connect nodes to the bus cable.
A drop line is the part of the main cable that links to the device. A tap is a connector
that establishes contact with the metallic core of a cable by either splicing into the
main cable or piercing the sheathing of a cable. As a signal travels along the backbone,
it loses some energy and transforms into heat. Consequently, as it travels further, it
weakens. Due to this, a bus can only accommodate a specific number of taps and a
certain distance between taps.

Figure 1.3. Functionality of bus topology (Source: Computernetworks969, https://i0.wp.com/www.

studytonight.com/computer-networks/images/Figure5.png).

a. Features of Bus Topology

1. It transmits data only in one direction.
2. Every device is connected to a single cable
b. Advantages of Bus Topology
1. It is cost effective.
2. Cable required is least compared to other network topology.
3. Used in small networks.
4. It is easy to understand.
5. Easy to expand joining two cables together.
c. Disadvantages of Bus Topology
1. Cables fails then whole network fails.
2. If network traffic is heavy or nodes are more the performance of the network
decreases.
3. Cable has a limited length.
4. It is slower than the ring topology.
10 COMPUTER NETWORKS

1.1.3.2. RING Topology

Each device in a ring topology has a unique point-to-point connection
with just the other two devices on either side. From device to device,
a signal is transmitted in a single direction around the ring until it
reaches its target. Every gadget in the ring has a repeater built in.
A device’s repeater regenerates the bits and transmits them when
it receives a signal meant for another device.

Figure 1.4. Structure of ring topology (Source: Geeksforgeeks, https://

media.geeksforgeeks.org/wp-content/uploads/20200526102238/Untitled-
Diagram-153-1.png).

a. Features of Ring Topology

1. For a ring topology with 100 nodes, many repeaters are
Remember needed because, in order to get to the 100th node, data
Computer must travel through 99 nodes. This is why in large ring
networking is the topologies with lots of nodes, repeaters are used. Therefore,
branch of computer the network uses repeaters to prevent data loss.
science that deals
with the ideation, 2. The transmission is unidirectional, but it can be made
architecture, bidirectional by having 2 connections between each Network
creation, Node; it is called Dual Ring Topology.
maintenance,
and security of 3. In Dual Ring Topology, two ring networks are formed, and
computer networks. data flow is in the opposite direction in them. Also, if one
It is a combination ring fails, the second ring can act as a backup to keep the
of computer
network up.
science, computer
engineering, and 4. Data is transferred in a sequential manner, bit by bit. Data
telecommunication. transmitted has to pass through each node of the network
until the destination node.
b. Advantages of Ring Topology
1. Transmitting network is not affected by high traffic or by
adding more nodes, as only the nodes having tokens can
transmit data.
 COMPUTER NETWORKS AND SERVICES 11
2. Cheap to install and expand Optical Fiber or coaxial cable.
c. Disadvantages of Ring Topology b. Advantages of Star Topology
1. Troubleshooting is difficult in ring 1. Fast performance with few nodes
topology. and low network traffic.
2. Adding or deleting the computers 2. Hub can be upgraded easily.
disturbs the network activity. 3. Easy to troubleshoot.
3. Failure of one computer disturbs 4. Easy to setup and modify.
the whole network.
5. Only that node is affected which
has failed, rest of the nodes can
1.1.3.3. STAR Topology
work smoothly.
Every device in a star topology has a c. Disadvantages of Star Topology
dedicated point-to-point link connecting it
to the hub or central controller. There’s 1. Cost of installation is high.
no direct connection between the devices. 2. Expensive to use.
Unlike mesh topologies, star topologies 3. If the hub fails, then the whole
don’t allow direct communication between network is stopped because all the
devices. When a device needs to send data nodes depend on the hub.
to another, it goes through the controller,
4. Performance is based on the hub;
which then forwards it to the other
that is, it depends on its capacity.
connected device. That’s how the controller
acts as an intermediary.
1.1.3.4. MESH Topology
The link to other nodes or devices is point-
to-point. There is a connection between
every network node. Mesh can link n devices
with its n(n-1)/2 physical channels. There
are two techniques to transmit data over
the Mesh topology: they are:
1. Routing; and
2. Flooding.

a. MESH Topology: Routing

Figure 1.5. Arrangement of star topology (Source:
BYJU’S, https://cdn1.byjus.com/wp-content/ According to network specifications, each
uploads/2022/07/star-topology.png). node in routing has a routing logic similar
a. Features of Star Topology to how routing logic guides data to take the
shortest path to its destination. Alternatively,
1. Every node has its own dedicated routing logic that avoids those nodes, etc.,
connection to the hub. based on information about broken links.
2. Hub acts as a repeater for data For the purpose of reconfiguring the failed
flow. nodes, routing logic is even an option.
3. Can be used with twisted pair,
12 COMPUTER NETWORKS

b. MESH Topology: Flooding

Since flooding transfers the same data to every network node,
routing logic is not needed. Since the network is strong, data loss
is extremely unlikely. However, it causes unnecessary strain on the
network.

Figure 1.6. Networking of mesh topology (Source: Wikipedia, https://

upload.wikimedia.org/wikipedia/commons/thumb/9/97/NetworkTopology-
Mesh.svg/800px-NetworkTopology-Mesh.svg.png).

a. Types of Mesh Topology

1. Partial Mesh Topology: In this topology, some of the systems
are connected in the same fashion as mesh topology, but
Remember some devices are only connected to two or three devices.
Mesh networks 2. Full Mesh Topology: Each and every node or device is
offer high connected to each other.
redundancy
because there are b. Features of Mesh Topology
multiple paths
for data to travel 1. Fully connected.
between any pair 2. Robust.
of nodes. If one
link or node fails, 3. Not flexible.
data can still be c. Advantages of Mesh Topology
routed through
alternative paths. 1. Each connection can carry its own data load.
2. It is robust.
3. Fault is diagnosed easily.
4. Provides security and privacy.
d. Disadvantages of Mesh Topology
1. Installation and configuration is difficult.
2. Cabling cost is more.
3. Bulk wiring is required.
 COMPUTER NETWORKS AND SERVICES 13
1.1.3.5. TREE Topology
It has a root node and all other nodes are connected to it forming a hierarchy. It is
also called hierarchical topology. It should at least have three levels to the hierarchy.

Figure 1.7. Features of mesh topology (Source: Pinterest, https://static.javatpoint.com/tutorial/

computer-network/images/tree-topology-advantages-and-disadvantages.png).

a. Features of Tree Topology

1. Ideal if workstations are located in groups.
2. Used in Wide Area Network.
b. Advantages of Tree Topology
1. Extension of bus and star topologies.
2. Expansion of nodes is possible and easy.
3. Easily managed and maintained.
4. Error detection is easily done.
c. Disadvantages of Tree Topology
1. Heavily cabled.
2. Costly.
3. If more nodes are added maintenance is difficult.
4. Central hub fails, network fails.

1.1.3.6. HYBRID Topology

It’s a mix of two or more topologies and comes in two different varieties. A hybrid
topology (ring topology and star topology) is formed when two topologies are connected
in an office setting. For example, one department uses ring topology while another
department uses star topology..
14 COMPUTER NETWORKS

Figure 1.8. Structures of hybrid topology (Source: JavaTpoint, https://static.javatpoint.com/tutorial/

computer-network/images/tree-topology-advantages-and-disadvantages.png).

Did you know?

In the 1970s and 1980s, Ethernet emerged as a popular networking technology developed by Xerox PARC,
which later became a standard known as IEEE 802.3. Ethernet initially used a bus topology, but it later evolved
to support various topologies, including the star topology.

a. Features of Hybrid Topology

1. It is a combination of two or topologies
2. Inherits the advantages and disadvantages of the topologies included
b. Advantages of Hybrid Topology
1. Reliable as Error detecting and troubleshooting is easy.
2. Effective.
3. Scalable as size can be increased easily.
4. Flexible.
c. Disadvantages of Hybrid Topology
1. Complex in design.
2. Costly.

1.1.4. Unique Identifiers of Network

A string of numbers or alphanumeric characters linked to a single entity inside a system
is called a unique identifier, or UID. That entity can be addressed using a UID in order
to be accessed and interacted with. Anything that has to be identified differently
from other entities, including machines, websites, businesses, and individual users,
can be given a unique identifier. These unique values can be allocated incrementally,
arbitrarily generated by an algorithm, or selected by the user. The assignment of these
 COMPUTER NETWORKS AND SERVICES 15
distinctive values is typically contingent upon the requirements of
the particular application.
The process through which users register for a website or
service is the most well-known application of unique identifiers.
In order to distinguish themselves in user logs, customers are
frequently given a username or user ID by the business they are
registering with. Afterwards, security and log-on uses are made
of these identifiers.
In a physical supply chain, UIDs are also commonly used.
Manufacturers often attach a serial number to the entire product or
specific parts of a larger component, such as computer parts. This
practice allows consumers to trace the product back to its original
manufacturer in case of issues like malfunctions, defects, or recalls.
The healthcare sector also makes frequent use of unique identifiers.
Instead of the patient’s name, a personal code is generated for
reporting medical information. This approach not only addresses
patient privacy concerns but also eliminates the possibility of
duplicate reports. Regarding host names, a hostname is a unique
device name assigned to each device in a network. To view your
machine’s hostname, simply type “hostname” in the Administrator
Mode command prompt and press “Enter.”

Keyword
Network
interface
card (NIC)
is a hardware
Figure 1.9. Unique identifiers of network (Source: Geeksforgeeks, https://
component,
media.geeksforgeeks.org/wp-content/uploads/hostname.png). typically a circuit
board or chip,
• IP Address (Internet Protocol address): The system’s installed on a
network address over the network is also referred to as computer so it
can connect to a
the Logical Address. The Internet Assigned Numbers network.
Authority (IANA) assigns an IPv4 (Version 4) address to
each device on the Internet so that it can be uniquely
identified in the global web. The IP address has a length
of 32 bits. Consequently, 2^32 IP addresses are available
16 COMPUTER NETWORKS

to us. In the command prompt, type “ipconfig” and hit “Enter” to obtain the
device’s IP address.
• MAC Address (Media Access Control address): The NIC (Network Interface
Card) is linked to the unique identifier of every host, which is also referred to
as the physical address. At the time of manufacture, the NIC is given a MAC
address. The MAC address has the following length: 12 nibbles/6 bytes/48
bits. Type “ipconfig/all” into the command prompt and hit “Enter” to obtain
the MAC address.
• Port: A port is a logical channel that an application can use to send and receive
data. Numerous applications can be running on a single host, and the port
number of each application allows for its identification. Since port numbers
are 16-bit integers, there are 216 ports available, which are divided into the
following categories:

PORT TYPES RANGE

Well known 0 – 1023

Ports
Registered Ports 1024 – 49151

Ephemeral Ports 49152 – 65535

Number of ports: 65,536 Range: 0 – 65535

Type “netstat -a” in the command prompt and press ‘Enter’, this lists all the ports
being used.

Figure 1.10. Identification of active network connection (Source: Geeksforgeeks, https://media.

geeksforgeeks.org/wp-content/uploads/ports.png).

• Socket: The unique combination of IP address and Port number together are
termed as Socket.
• DNS Server: The domain name system is known as DNS. In essence, DNS is
a server that translates URLs, or web addresses (e.g., www.google.com), into
 COMPUTER NETWORKS AND SERVICES 17
the IP addresses that match them. It is not necessary for us to retain the IP
addresses of every single website. You can find the IP address of the domain
you’re looking for by using the ‘nslookup’ command. This also provides details
about our DNS server.

Figure 1.11. Translation of web addresses or URLs (ex: www.google.com) into corresponding IP
addresses (Source: Geeksforgeeks, https://media.geeksforgeeks.org/wp-content/uploads/nslookup.
png).

• ARP: Address Resolution Protocol is what ARP stands for. It is employed to

translate an IP address into the matching physical address (i.e., MAC Address).
The Data Link Layer uses ARP to determine the Receiver’s machine’s MAC
address.
• RARP: Reverse Address Resolution Protocol is what RARP stands for. As the
name implies, it accepts a physical address as input and outputs the device’s IP
address. However, since DHCP entered the scene, RARP has become outdated.

Did you get it?

1. Differentiate the functionality of bus topology and hybrid topology.
2. What are the unique identifiers of network?
1.2. TYPES OF
NETWORK SERVICES
→Learning Objectives

• Examine the strategy of user management

• Know about the email management
• Find the process of printer sharing
• Give an overview on system administration

A vast array of connectivity and software tools that are distributed to computers
connected to a network and overseen by a central team are collectively referred to as
network services. When several computers are connected to a central server or to each
other, the result is a networked computer environment. Shared files and utilities are
accessible from a central location to all computers. This kind of environment has many
benefits, from improved quality control to performance and issue management. System
administration, email, printing, and user management are the four categories of network
services. Desktop computers are referred to as client computers or workstations in a
networked environment.
The servers are normally kept in a separate, temperature-controlled room that is
frequently housed in a different structure than the areas where the workstation users
are situated. Network cables, which are used to transfer data packages within the
network, are used to provide access. Additionally, these cables offer access to the
Internet, usually following a path via a network switch or central server. A variety of
functions are included in user management, such as assigning rights, privileges, and
access in addition to creating user names and passwords.
Email is a widely used tool for productivity that typically requires an Internet
connection. Policies in many organizations govern the email software, maximum storage
limits, and file sizes for transmission. Most companies provide their employees with
company-specific email addresses, which necessitates the setup and maintenance of
an email server. Shared printing is one of the key network services required in any
organization. All users can send printing tasks to a central unit, eliminating the need

18 COMPUTER NETWORKS
 COMPUTER NETWORKS AND SERVICES 19
for individual printers at each desk. This device often serves as a
multifunction printer, scanner, and copier. Many businesses utilize
specialized printer management software to handle document flow Keyword
and protect sensitive information. System
administration
An essential network service for system administration is the is the field of work
in which someone
ability to manage all workstations from a single location. The manages one or
administrator is responsible for installing new software, updating more systems,
operating systems, and ensuring all workstations run the same be they software,
hardware, servers
software versions. In many businesses, workstations are locked, or workstations.
granting only the system administrator permission to install updates
or launch new applications. This practice reduces staff time spent
on computer maintenance and enhances the overall quality of the
network.

1.2.1. User Management

Effectively managing users and their accounts to grant them
access to a range of IT resources, including storage systems,
networks, devices, apps, systems, and SaaS services, is known
as user management, or UM. Administrators can grant access
and take control of user accounts through user management, also
known as user access management. A user management system is
a fundamental security tool that is a crucial component of identity
and access management (IAM).
In an organization’s IT infrastructure, user management
enables administrators to add, edit, and remove user accounts.
An administrator can establish a user account for a newly hired
employee, outlining the user’s role, permissions, and access rights,
for instance. The administrator grants the necessary permissions
when a user needs access to particular resources, like files or
applications. These rights specify what an individual is allowed to
do on the network of the company and what they are not. A sales
representative, for example, might only have access to customer
information pertinent to their position, whereas a marketing manager
might have access to marketing tools and data.
User management simplifies the process of granting and
controlling access to IT resources overall. It guarantees that
users have the rights to carry out their jobs efficiently while
upholding security and compliance standards inside the company.
An individual’s access to devices, software, and services can be
managed through a system called user management. Its main
objectives are to monitor usage and manage permissions for actions
and access.
20 COMPUTER NETWORKS

Functions of user management include:

• Providing users with authenticated access
Remember • Supporting set up, reissuing, and decommissioning of users’
User access credentials
management is a
system to handle • Establishing access privileges based on permissions
activities related Accounts connected to software licenses can also be monitored
to individuals’
by user management at every stage of their existence. This makes
access to
devices, software, sure that everyone using the software has licenses for it, and that
and services. these can be reclaimed and reissued when they’re not in use.
It focuses on
managing
permissions
for access and
actions as well
as monitoring
usage.

Figure 1.12. Administrators to manage users and the actions they are
permitted to take on objects.

Source: https://www.miniorange.com/images/user-management/user-
management.webp

User Management and the Cloud

User management in cloud applications and resources calls for
increased caution. To handle the spread of users and the proliferation
of accounts, IT departments must develop and oversee more intricate
policies. IT teams also have to keep track of the different user
management systems that the cloud service providers use, which
complicates this already difficult task. This is due to the fact that
user management in the cloud varies based on the service provider
and the kind of deployment. Here are two popular approaches to
cloud user management:
1. Identity and Access Management (IAM): The technology,
regulations, and procedures gathered here are employed to
oversee user accounts and associated access (i.e. e. humans)
and functions (i.e. e. scripts or services). IAM is connected
to roles or users. Permissions based on a user’s job are
 COMPUTER NETWORKS AND SERVICES 21
defined using a system called Role- By monitoring users and their
Based Access Control, or RBAC. usage, user management software
2. Resource Access Management helps organizations make sure
(RAM): Policies specify who can license agreements are followed.
access a resource and what actions Additionally, this makes reporting
can be performed. RAM is tied to easier in the event of an audit.
the resource rather than to a user • Security benefits with user
or role. management software
Security benefits from user
Benefits of User Management management software are
substantial. Unauthorized access
User management software can help
can be avoided through the support
organizations gain productivity, security,
of stringent access controls. Further
and cost savings.
reducing insider threat is the
• Productivity benefits with user speed with which systems can be
management software swiftly locked down or removed.
Software-enabled user management Root cause analysis, remediation
automatism boosts productivity in the event of a data breach, and
and reduces waiting times by forensic audits are further features
mirroring modifications (e.g., supported by user management
adding, changing, and deleting software.
users) between various platforms.
Moreover, it streamlines the process User Management—Benefits from Security
of creating users, roles, and groups, and Beyond
which lessens the administrative Software tracking and management has
teams’ workloads. always been a little tricky. This complexity
• Cost-savings benefits with user has only grown as a result of the widespread
management software use of cloud applications, which may have
Software usage can be tracked also made user management more crucial.
with the help of user management Organizations are more vulnerable to
software to guarantee the best possible data breaches as user anonymity
possible licensing. It is possible to increases and perimeters become more
reassign licenses that are no longer permeable. In order to prevent internal and
required. Software agreements external users from gaining unauthorized
that are no longer required may access to files, applications, systems, and
be terminated. For organizations devices on-premises and in the cloud, IT
looking to optimize license can maintain control over users’ activities
distributions, knowing how many and reinforce other security measures
devices a user has activated under with the aid of user management. One
their license is helpful. It also aids advantage of user management is that its
in budgeting for software in the advantages go beyond IT defense, which is
future. not the case with other security solutions.
Through optimized license usage, it lowers
• Software license compliance
costs, which has an impact on finance.
benefits with user management
software
22 COMPUTER NETWORKS

Lastly, it helps with the onboarding and packages. Although client software is still
decommissioning of corporate identities for in use today, a more adaptable approach
contractors and employees, which benefits has gained popularity recently. Webmail is
IT operations and human resources (HR). another well-liked method of accessing email
in addition to client software. As an extra
bonus, the majority of service providers
1.2.2. Email Management allow users to access their mailboxes via the
Email is just the abbreviated version of Internet from any computer or other web-
“electronic mail,” which is a system for connected device. There are email services
sending, receiving, and storing electronic that are made especially to use the Internet
messages. It is also sometimes written as as their main interface. These websites,
e-mail. With the growth of the Internet, its known as “portals,” have gained a lot of
popularity has increased almost universally popularity because they typically allow
worldwide. Email is now frequently used as users to register for a free email account
the go-to form of communication for both and provide access to a calendar, news,
personal and professional matters. search engine, and many other resources
from one location.
Email messages typically arrive in the
recipient’s account in a matter of seconds. Although the exact date of the
Images and various types of formatted invention of email is somewhat unknown,
documents are now easily included as Ray Tomlinson, an American, is widely
attached files, so they usually contain recognized as the father of the modern
more than just text. Furthermore, sending version. Prior to Tomlinson, users could
or receiving emails no longer requires a communicate with one another via messages,
computer in front of you. Various mobile but only if they had the same computer
devices, including tablets and smartphones, connected. Not even after computers were
enable correspondence management while networked could messages be directed to a
on the go. Emails can move easily across specific user. Tomlinson was credited with
different kinds of servers on different one of the most significant inventions in
networks and between computers running communication of the 20th century because
different software thanks to standardized he came up with a method to address mail
protocols that users use to send and receive to specific users. Tomlinson came up with
messages. SMTP, or simple message transfer the idea of identifying the user’s name
protocol, makes it possible for messages to and the computer that they were using.
be sent and received. Users can retrieve and Because of this, the standard email address
store messages over time using additional format was username@usercomputer. With
protocols like Internet Message Access the exception of the user computer being
Protocol (IMAP) and Post Office Protocol frequently substituted with the name of
(POP). a service provider, this standard has not
altered significantly over time.
In the beginning, processing mail
required a program known as “client Legally speaking, email is typically
software.” Email clients come in a variety regarded as a private communication
of forms; some are free to use, some medium, similar to phone conversations or
require payment, and some Internet service even handwritten letters. While specifics
providers even bundle them into their may differ depending on the country, it
is usually prohibited to access, read, or
 COMPUTER NETWORKS AND SERVICES 23
publish someone else’s messages without
authorization. Nonetheless, it is crucial to Did you know?
realize that hackers can easily intercept In 1994, Information Cybernetics, a company in
electronic mail. Furthermore, a lot of Cambridge, Massachusetts, developed tools for pattern
businesses demand that workers sign a analysis and categorization of emails and other electronic
communication channels. The platform of tools was
consent form authorizing the business to
called EchoMail. The first company to adopt EchoMail
read and keep track of all emails sent and was AT&T. JC Penney adopted EchoMail in 1997.
received from company email addresses.
The fact that every electronic message
becomes a virtually permanent record of
1.2.2.1. Management Procedure
the conversation raises additional privacy
concerns. Currently, email is the de facto standard
for business communication within
When a message is marked for deletion,
organizations. The organizational standards
it is taken out of the user’s account’s
pertaining to information and records must
visible queue; however, deleted emails are
include email as well as adherence to it, just
infamously recoverable, so one should never
like any other form of business information
assume that the message is permanently
and record. Email is now extremely important
lost. Emails are likely to remain in existence
to business and is used in every part of all
on both the recipient’s end and many of the
organizations. Information workers spend
servers they traversed, even if the sender
many hours a day reading, replying to,
uses specialist software to completely
and working together via emails, with the
remove all traces of the message from their
email client serving as their main business
computer. As such, it is foolish.
application. Many people’s definition of email
Email’s impact is too great to be management is as simple as taking emails
understated. For instance, the USPS reports off a server and putting them in a repository.
that it processes 485 million pieces of mail However, in order to properly handle it,
every day. Comparatively, in the US, 39.6 all of the other types of documents and
billion emails are sent each day. Stated records must also be categorized, stored,
differently, the volume of electronic mail and possibly destroyed in accordance with
handled by providers surpasses 81 times established business policies and standards.
that of the world’s largest postal system. The methodical regulation of the volume
Examining how much the use of traditional and caliber of electronic messages sent and
“snail mail,” as it is now commonly called, received by an organization is known as
has been impacted by electronic mail may email management.
be equally instructive.
Email servers within the organization
Beginning in the latter half of the are negatively impacted by the large
1990s and continuing steadily ever since, amounts of emails stored in sent, deleted,
the postal service, like many other postal and inboxes, which also pose a risk to
services around the world, has seen the organization. The purpose of email
dramatic declines in volume and revenue; servers was never to store so many emails
this is partially due to the increasing amount and transfer control of that data outside
of personal and business correspondence of the company. A company’s effort and
that is now delivered by email. expenses when responding to eDiscovery
and disclosure will increase if it is unable
24 COMPUTER NETWORKS

to manage its email system and meet its more of a number of unacceptable options
legal preservation obligations in the event that they think constitute management but
of litigation or government investigations. really don’t. Organizations may decide the
best way to manage email is by doing one
Emails sent and received by staff or more of the following:
members are centrally recorded by email
management systems. Retention periods • Saving all email messages forever
and access controls can be used to manage • Saving all email messages in the
emails in addition to using a classification messaging application
scheme to manage this content. To manage
• Setting arbitrary mailbox sizes for
and retrieve this information, metadata
all users
related to the emails can be collected. One of
the most popular uses for email management • Declaring “email’ as a record series
is email archiving. These solutions, in their • And, of course, the ever-popular
most basic form, copy or delete messages “Doing nothing”
from the messaging app and store them The best answer is to ensure your
elsewhere. Some of these solutions archive organization has strong policies in place
the attachments but leave the messages to govern such things. And yet, we know
unarchived, replacing them with a stub or full well that not every organization does.
a link from within the message. Systems for According to a 2009 AIIM study on the
managing emails are a component of a larger subject:
solution. Records management programs,
which include personnel, activities, and • Only 10% of organizations have
policies and procedures applicable to the completed an enterprise-wide email
entire organization, are essential. management initiative,
Email is an information type that is often • Only 20% currently are rolling out
treated differently than others even though a project, and
it shouldn’t be. The reasons for different • Even in larger organizations, 17%
treatment are many and reflect: have no plans to do so.
There is much that organizations need
• The sheer volume of messages
to determine:
involved, which dwarfs the number
of other document types in play on • When to declare an email, copy
a daily basis of a message or an entire thread/
conversation a record
• The informality with which they
can be created and forwarded: • Determine what to base the record
everybody does it, all the time, decision upon – sender, receiver,
with the simple click of a mouse type of content, attachments, text
within the email, etc.
• The ease with which documents
can be attached, even if they’re
not supposed to be. 1.2.3. Printer Sharing
These factors make it difficult to The process of granting access to one
fully understand email management as a or more printers to numerous computers
discipline and wrestle it to the ground. and devices linked to the same network
Organizations will often default to one or is known as printer sharing. Depending
 COMPUTER NETWORKS AND SERVICES 25
on the permissions the administrator has set for each user, every
node or device on the network can print to any shared printer and,
to a limited extent, modify the printer settings. A computer can
share a printer with other computers on the same network if it is
connected to a printer that supports printer sharing. As long as the
shared printer is set up correctly on one computer, that computer
can share it regardless of how old or new it is. The operating Keyword
system (OS), which manages communication between computers and Operating
System (OS)
devices on the network as well as the printer itself, makes sharing is a collection
possible. The computer connected to the shared printer receives of software that
print requests from other networked computers and uses them to manages computer
hardware resources
initialize the printer before forwarding print jobs to it. Regretfully, and provides
the person initiating the print job still needs to manually retrieve common services
the actual printouts. for computer
programs.
To enable printer sharing in Windows 7 and 8, navigate to the
Network and Sharing Center icon, click “Change advanced sharing
settings,” and then select the “Turn on file and printer sharing”
radio button under “File and printer sharing.” Each computer (Client
PC) in a network shared environment sends data via a server PC,
which is managed centrally, as shown below. This kind of computer
is frequently referred to as a “Print Server” or “Server.”
In a Network Shared environment:
• It is possible to use both print and scan functions from a
server PC, but only print from client PCs. (Note: Only if
your Brother machine has a scan function.)
• Client PCs cannot use the shared printer when a server PC
is turned off.

1.2.3.1. Peer-to-Peer
Every computer sends and receives data to every device directly
in a peer-to-peer setting. File access and machine sharing are not
controlled by a single central server. We recommend using the peer-
to-peer printing method, as shown below, if your Brother device is
compatible with wired or wireless Ethernet networks.
• In a Peer-to-Peer environment,
• It is easier to configure than the Network Shared printing
method.
• Client PCs can print to the printer at any time.
• No need to be concerned if the Server PC is turned on or
not.
• You can use the print and scanning functions of the Brother
26 COMPUTER NETWORKS

machine. The functions available on Access control, user management,

the Peer-to-Peer connection vary backup and recovery, and system
depending on the machine. For monitoring are among the duties of system
details, we recommend you check administration. Usually, backup, recovery,
the machine specification. and system monitoring features are
combined into a single application for the
1.2.4. System Administration entire company. Role assignment and user
creation are examples of user management
Managing a hardware and/or software system functions. Using the same login ID and
is known as system administration. A system password to access various systems within
administrator is responsible for handling an organization is known as single sign-on, or
various tasks such as backups, granting user SSO. Software libraries frequently integrate
access, managing user accounts, ensuring authorization and authentication into a single
system security, monitoring system health, component. A computer network’s overall
and allocating system resources like disk functionality, including network security, the
space. IT professionals typically work on individual PCs and hardware connected to
system administration for an organization, the network, as well as any software the
ensuring that computer systems and computers may be running, are all under the
associated services operate smoothly. The control of computer system administrators.
responsibilities of system administrators are Computer system administrators must
diverse and can vary based on the types of regularly upgrade hardware and software,
computer systems they maintain, but they troubleshoot any issues that arise during
often involve common tasks that can be upgrades, and consult with staff regarding
approached in different ways. upcoming system upgrades in order to
Tasks like installing new hardware or achieve the goal of satisfactory system
software, setting up and maintaining user functioning. Nowadays, the majority of
accounts, managing computer systems such system administrator jobs demand at least
as servers and databases, and preparing for a bachelor’s degree in computer science,
and addressing system outages are common. with US and larger corporations setting
Additional duties may include providing the standard.
guidance to computer users, assisting with When it comes to system administrator
light programming or scripting to streamline tools, computer users are more likely to
system workflows. While having some think of physical, tangible items like cables,
programming knowledge can be beneficial, power cords, and various analyzers that are
system administration is not primarily intended to measure and analyze signal
a programming role. Modern Microsoft problems and system efficiency than people
Windows systems often use PowerShell in the industry. Punchdown tools, sometimes
for detailed system information, while referred to as punch-down tools or Krone
Linux and UNIX systems offer a variety of tools, are the only tools that a system
programming languages like Bash, Perl, and administrator can truly use that resemble
Python. While proficiency in these languages tools. Wires are inserted into tiny spaces
is not mandatory, reaching an intermediate inside computers using punchdown tools.
level can enhance your understanding of Software created for various purposes is also
system operations and improve efficiency included in the system administrator tools.
in daily tasks. System administrator software, sometimes
 COMPUTER NETWORKS AND SERVICES 27
referred to as admin source tools, is not visible kinds of software support safe business
to end users, in contrast to the software transactions, preserve data privacy, and
utilized on network user PCs. Rather, it provide. Many of the tasks that once
provides the system administrator with a required continuous attention from network
variety of data, including software usage system administrators are now automated
trends, computing speeds across different and are regarded as essential administrator
programs and databases, and software tools of the trade. Numerous native tools,
updates that are installed automatically but also referred to as system administrator
need the system administrator’s approval. tools, come pre-installed with the software
System security-related software is another that they are meant to manage. Third-party
category of software that is deemed vendors sell other kinds of automated
necessary for system administrators. These software made to make system administrator
tasks easier.

Did you get it?

1. What are the useful factors for user management?
2. Determine the management procedure of email networking.
1.3. TYPES OF NETWORK
→Learning Objectives

• Understand the local area networks (LAN)

• Describe the personal area networks (PAN)
• Know about the home area networks (HAN)
• Learn about the wide area networks (WAN)
• Discuss on campus networks
• Give an overview on metropolitan area networks (MAN)
• Examine the enterprise private networks
• Define the purpose of internetworks
• Define the backbone networks (BBN)
• Explain the purpose of global area networks (GAN)

A computer network is an assembly of interconnected computers that facilitates computer-

to-computer communication and resource, data, and application sharing. A collection of
two or more linked computer systems is called a computer network. A network connection
can be made with wireless or cabled media. Hardware and software are used in every
network to link computers and other devices.
A computer network is a type of digital communications network that permits resource
sharing between nodes. Computer networks use connections (data links) between nodes
to facilitate data exchange between computing devices. These data links are set up
using wireless technologies like Wi-Fi and cable technologies like twisted pair or fiber-
optic cables.
Network nodes are computer systems on a network that start, route, and end data.
Network addresses are typically used to identify nodes, which can include networking
hardware like switches and routers as well as hosts like PCs, phones, and servers. When
two of these gadgets can communicate with each other, regardless of whether they
are directly connected, it is considered that they are networked together. Application-
specific communications protocols are typically layered (i.e., transported as payload)

28 COMPUTER NETWORKS
 COMPUTER NETWORKS AND SERVICES 29
via additional, broader communication protocols. To maintain the
dependability of this formidable collection of information technology,
expert network management is needed.
A vast array of services and applications are supported by
computer networks, including access to the World Wide Web, digital Remember
video and audio, shared use of printers, fax machines, application Computer
and storage servers, email and instant messaging, among many networks may
other things. Different computer networks have different bandwidths, be classified by
transmission media for carrying their signals, communications many criteria,
protocols for managing network traffic, sizes, topologies, traffic including the
transmission
control mechanisms, and organizational goals. The Internet is the medium used
most well-known computer network. to carry signals,
bandwidth,
There are many different types of networks, which can be communication
used for different purposes and by different types of people and protocols to
organizations. Here are some of the network types that you might organize network
traffic, the
come across: network size,
the topology,
traffic control
1.3.1. Local Area Networks (LAN) mechanisms, and
organizational
A local area network, or LAN, is a collection of peripherals and intent.
computers that are connected to a server within a specific geographic
area via a wireless link or shared communications line. As few as
two or three users in a home office or several hundred users in a
corporation’s central office can be connected to a local area network.
In order for network nodes to share resources like printers or
network storage, LANs are set up by homeowners and information
technology (IT) administrators. Cables, switches, routers, and other
hardware are needed for LAN networking in order to connect
users to internal servers, websites, and other LANs that are a part
of the same wide area network (WAN). The two main methods
for making LAN connections available are Ethernet and Wi-Fi. A
specification called Ethernet makes it possible for computers to
talk to one another. Wi-Fi connects computers to LANs via radio
waves. While Ethernet and Wi-Fi speeds increased and connectivity
costs decreased, other LAN technologies, such as Token Ring, fiber
distributed data interface (FDDI), and ARCNET, fell out of favor. A
wireless LAN’s cost and flexibility make it potentially better than
a wired LAN connection in many circumstances. People who work
exclusively with smartphones, tablets, and other mobile devices are
common among companies that view WLANs as their main source
of connectivity. Virtual LANs, which enable network managers
to logically divide and group network nodes without requiring
significant infrastructure changes, were made possible by the trend
toward virtualization.
30 COMPUTER NETWORKS

Networking is integrated into many of networked devices without a centralized

operating systems, such as Apple OS X network system is called a personal area
and Microsoft Windows. This means that network. These networks are frequently
connecting computers in a network is not built with the aid of a mobile computing
too challenging, as long as the network device, like a laptop or smartphone. It
administrator has a fairly modern laptop or doesn’t really matter how the devices are
desktop PC. To set up a wireless network, connected, so Bluetooth™ is just as likely
the administrator will need an Ethernet to be used in a personal area network as
cable to connect the wireless router to the Wi-Fi™ or network cables. A personal area
main PC or server, as well as a wireless network connecting to the Internet was not
router connected to a broadband connection. common in the past, but with the increasing
This setup will allow other computers with popularity of Wi-Fi™ and mobile Internet,
built-in or connected wireless networking that has changed.
hardware to detect wireless signals and
join the local area network. A local area network (LAN) is a type
of network that is commonly constructed
using a centralized routing system. These
1.3.1.1. Characteristics of a LAN central systems, which are typically routers,
establish wired or wireless connections
LAN sizes can vary quite a bit. A local area
with each device on the network. The
network (LAN) consists of devices connected
router is used by the system’s devices to
through a home internet connection. In
communicate with one another. The priority
small businesses, LANs connect around a
and sequence of transferred traffic are also
dozen to a hundred PCs with printers and
set by the router. Finally, the router is the
file storage. The largest LANs are managed
means by which the LAN communicates
by a server that handles file storage, routes
with the Internet. Building a personal area
files to printers and scanners, and shares
network involves some slight variations. All
data among devices. The devices on a LAN
connected devices typically connect to one
are typically located in the same building
main hub, though daisy-chaining devices
as the network, distinguishing it from other
together is an option if the hardware
networks like the internet. These devices—
permits it. The linked devices connect to
computers, printers, scanners, and more—
one another through this central device.
connect wirelessly to a router through a Wi-
After that, they simply function as regular
Fi access point or via an Ethernet cable. It’s
extensions of the central device, managing
possible to connect multiple LANs together
movement and operation priorities in the
using radio waves or phone lines.
same manner as any of its regularly installed
systems.
1.3.2. Personal Area Networks The building of the network takes
(PAN) precedence over the techniques utilized
A network built around a person’s workspace to establish a personal area network. As
is called a personal area network. The a result, there are numerous standard
device owned by the individual serves as techniques utilized in their construction.
the hub of the network, to which other The practice of wiring computers together
devices are connected. Wireless personal was widespread in the past, but with
area networks are another option. A group the advent of wireless connections, it has
become less common. Certain devices in the
 COMPUTER NETWORKS AND SERVICES 31
network may use Bluetooth in addition to wireless technology. The
effective broadcast distance of a personal area network’s individual
components is the only real restriction, aside from those resulting
Keyword
from construction. The effective broadcast range of Wi-Fi TM and Network cabling
is the physical
BluetoothTM is approximately 30 feet (10 meters). This indicates that infrastructure that
two of the pieces could be as much as 60 feet (20 meters) apart connects computers
under normal circumstances. It is possible to expand that even and other devices to
a network.
further if the device supports multiple connections.

1.3.3. Home Area Networks (HAN)

Home environment devices are connected by a home area network.
Personal computers, iPads, iPhones, printers, TVs, and other gadgets
may be included. The Home Area Network, or HAN, is a network
that makes it easier for devices in close proximity to a home to
communicate with one another. An internal network used for energy
monitoring and control is called a home area network, or HAN. It
monitors energy consumption and connects to the electric meter of
the utility company so that the homeowner can see what energy
is being used and where. Additionally, it serves as a means for
the utility to manage the air conditioning and heating in order to
conserve energy. Moreover, the HAN offers a practical method for
turning appliances on and off. A home area network, or HAN for
short, is a kind of network that enables multiple computers and other
network devices in a room to share resources, including internet
connectivity. Any device with a TCP/IP connection can access the
internet these days thanks to a router installed in practically every
home. This covers the standard desktop or laptop, as well as the
newest home appliances (IoT), smartphones, smart TVs, gaming
consoles, and security cameras.
A typical architecture of a Home Area Network:
Internet service providers provide fiber-optic networks that are
directly connected to homes, offering lightning-fast VoIP, gaming, and
TV streaming services. A TCP/IP connection between the Internet
Service Provider and the home router is used for all communications.
In order to guarantee complete Wi-Fi signal coverage inside the
house, we typically have one or more wireless access points.
• Computers such as desktops, laptops, netbooks, and tablets
• Network-attached storage (NAS) device
• Network Printers
• Smartphones connected via Wi-Fi
• Smart speakers
32 COMPUTER NETWORKS

• Smart TV’s and DVRs (access A third option for creating a network,
to services such as Netflix and ISDN (integrated services digital network)
YouTube) solutions, became very popular in the 1990s,
• Stereo systems with network primarily due to the drawback.
connectivity that allow a user to
easily access their music library Did you know?
stored on a PC or NAS WANs are often built using leased lines. At each
• Video game consoles for multiplayer end of the leased line, a router connects the LAN on
one side with a second router within the LAN on the
games and social network other. Because leased lines can be very expensive,
integration instead of using leased lines, WANs can also be built
• Security alarms using less costly circuit switching or packet switching
methods.
• Garage door and gate openers
• HVAC (keeping your house warm A wide area network, or WAN, enables
or cold) businesses to utilize shared resources to run
• Smoke/CO detectors their operations. For instance, many retail
pharmacies use one to assist consumers who
• Media Players or Streaming Devices
fill prescriptions at one of their locations.
like Apple TV
Once a customer is in the pharmacy’s
common customer database, they can fill
1.3.4. Wide Area Networks a prescription at any of the company’s
(WAN) locations—even when they’re on vacation
in a different state.
A wide area network extends over a greater
area of space, typically within a radius of Wide area networks are useful for
several kilometers. Wide area networks, businesses as well. With this kind of
or WANs for short, are communications wide area network application, internal
networks that connect local computer operations like sales, marketing, production,
networks to a wider, functional network development, and accounting can also be
that can span both domestic and foreign shared with approved locations. Employees
locations by utilizing current technology. can work from almost anywhere by using
The local area network and the metropolitan this to combine disparate location-based
area network, on the other hand, offer computer networks into a single computer
communication inside a constrained network for the entire company. In the event
geographic area. It is frequently desirable that a natural disaster damages or renders
to link computer networks, particularly one facility unusable, staff members simply
for companies that run multiple locations. relocate and continue working using the
Utilizing the current telephony technology shared network at their new location.
makes this the easiest to do, starting with
the local area network and progressing to
the wide area network. Fiber optics are 1.3.5. Campus Networks
essentially used to establish a connection A campus network is like a local area
between networks that are housed in network (LAN) used by government
various facilities. This usually entails using agencies, universities, businesses, and
PSTN (public switched telephone network) similar organizations. It usually covers a
technology or POTS (standard phone lines). group of nearby buildings. In a campus
 COMPUTER NETWORKS AND SERVICES 33
network, you’ll find a set of connected LANs exclusive to a business,
government body, academic institution, or similar establishment.
Typically, a campus includes a cluster of adjacent buildings. While
users in a campus network are not as spread out as in a wide area
network (WAN), they may be more geographically dispersed than in
a single LAN. Campus networks in colleges and universities connect
various buildings like administration offices, housing complexes,
lecture halls, libraries, student centers, sports facilities, and more
within a specific town or neighborhood. Important departments
and staff members work in buildings linked by corporate campus
networks. The corporate campus network, within a limited geographic
area, acts as the user-facing part of the larger corporate network.
In an ideal campus network, every node is connected to every other
node via optical fiber media, utilizing Gigabit Ethernet or 10-Gigabit
Ethernet technology. In certain places, such as university student
centers or libraries, where multiple users concurrently use portable
and mobile devices like notebook and tablet computers for research
and communication, Wi-Fi hot spots or even a hot zone comprise Keyword
the user end of the network. Occasionally, the phrase “campus Switched
network” refers to a group of geographically dispersed Internet Multimegabit
Data Service
users who share a common interest. Examples of such groups (SMDS) is a public,
include the International Sustainable Campus Network (ISCN), a packet-switched
forum that assists colleges, universities, and businesses in their service aimed at
enterprises that
pursuit of sustainability in research and teaching, and the Roosevelt need to exchange
Institute Campus Network, a nationwide student initiative. large amounts of
data with other
enterprises over a
wide area network
1.3.6. Metropolitan Area Networks (MAN) on a nonconstant or
Networks that cover an area the size of a metropolitan area are bursty basis.
known as metropolitan area networks. A Metropolitan Area Network,
or MAN, is made up of several interconnected Local Area Networks
(LANs) that may also be connected to a Wide Area Network (WAN).
A Metropolitan Area Network, or MAN, is a sizable computer network
that reaches across a city or a sizable university campus. Long-term
internet connectivity is the main goal of the Metropolitan Area
Network (MAN). Several LANs are typically combined into a MAN
(Metropolitan Area Network) to create a network. An optical fiber
setup makes up the backbone of this extensive network, known as
the Metropolitan Area Network (MAN).
The three technologies that are most frequently used to create
MAN (Metropolitan Area Network) networks are Switched Multi-
megabit Data Service (SMDS), ATM (Asynchronous Transfer Mode),
and Fiber Distributed Data Interface (FDDI). The most commonly
used is ATM (Asynchronous Transfer Mode). Digital data transfer
technology is known as ATM (Asynchronous Transfer Mode). It was
34 COMPUTER NETWORKS

created in 1980 with the goal of enhancing used to create a metropolitan area network,
real-time data transfer across a single or MAN. Routers and switches are used
network. Similar to a cell relay system, to set up the network. An active port that
ATMs (Asynchronous Transfer Mode) manages the filtering of data, typically in
divide data into fixed, equal-sized packets the form of frames, is called a switch. Every
that are transferred gradually. During a switch functions as two ports: on the one
video conference, the ATM (Asynchronous end, it handles data filtering, and on the
Transfer Mode) was used to access clear other, connection management. Another
audio and video results. ATMs’ qualities tool to help with network connectivity is
have made it possible for them to serve as a router. The data packets are assisted by
the foundation for wide area data networks. the router in determining the best course
of action. To put it another way, it monitors
Asynchronous Transfer Mode, or ATM, the transfer of data. An area of up to 50 km
combines packet switching and circuit is typically covered by MAN (Metropolitan
switching features to transfer data even Area Network).
in real time. The data transfer over LAN
standard FDDI has a range extension of The Metropolitan Area Network,
about 200 kilometers. Several thousand or MAN, is situated between the LAN
users’ data transmission can be supported and the WAN. As a result, it lowers the
with the aid of FDDI. It is known as cost associated with setting up a wide
the MAN (Metropolitan Area Network) area network while also improving data
technology for this reason. Fiber distribution handling efficiency. The Metropolitan Area
data interface, or FDDI, gets its name from Network, or MAN, provides centralized data
the fact that its fundamental infrastructure management. You can link numerous fast
is built on optical fiber. SMDS technology LANs together with it. Data transfer has
is used when transferring data via a been made easier by telephone companies
connectionless service. Data is transferred all over the world thanks to an underground
by storing information in the header and optical fiber network. The effectiveness and
traveling independently over any network speed of data transfer are improved by these
to reach its destination, according to optical fibers. A speed of nearly 1000 Mbps
connectionless services. Data transfers via is available to you thanks to the optical
SMDS technology result in the formation of fibers. A WAN with a speed of 1.45 Mbps
tiny data packets, much like those seen in will cost you more money than it will benefit
ATMs. Nevertheless, datagrams—unreliable you. On the other hand, a metropolitan area
data service provider’s data packets— network provides you with a total speed
can be transmitted over great distances of 1000 Mbps at the lowest possible cost.
thanks to SMDS. These days, microwave
and infrared signals are used to establish
MAN (Metropolitan Area Network) links. 1.3.7. Enterprise Private
Networks
Between LAN and WAN is typically
where MAN (Metropolitan Area Network) An organization connects its sites via an
is located. Connecting geographically enterprise private network, allowing the
separated LANs is the usual use for it. locations to share resources. An enterprise
In order to create a communication link private network is a type of computer
between two independent LAN nodes, network that enables large corporations
MAN’s objective. Optical fiber is typically with several dispersed offices to securely
 COMPUTER NETWORKS AND SERVICES 35
connect those offices to one another via a network. Sharing of
computer resources is the primary purpose of an enterprise private
network.
Enterprise Private Networks fulfill the original function of
private networks, which was to share resources while protecting
data. Businesses create enterprise private networks to link their
various locations, which include offices, manufacturing facilities,
retail stores, warehouses, and other company sites. In order to
exchange resources. Enterprise Private Networks, or EPNs, are a
cutting-edge illustration of how information technology benefits the Keyword
business sector. Businesses can easily expand their operations by
Public network
connecting various departments that are spread out across remote is a wireless or
locations and frequently do not have connectivity with one another wired network that
through a private network. computers and
devices you don’t
trust are connected
This has a wide range of effects because it significantly affects a to.
company’s ability to be sustainable. Different business divisions can
benefit from the increased connectivity and resource mobility that
comes with integrating an enterprise private network. Enterprise
private networks place a high value on privacy, which is typically
protected by a combination of security controls and tunneling
protocols like the Layer 2 Tunneling Protocol. Internet service
providers, or ISPs, use a type of computer networking protocol
called L2TP, or Layer 2 Tunneling Protocol. Such a protocol exists
only to facilitate private network functions, like VPN and EPN.
Although Layer 2 Tunneling Protocol is a session layer protocol, it
resembles the Data Link Layer Protocol found in the OSI reference
model quite a bit.
A User Datagram Protocol, or UDP, is included in the Layer 2
Tunneling Protocol to facilitate communication. IPSec, or Internet
Protocol Security, is used as an encryption protocol since the
original version of this protocol lacked any features that could have
maximized data security, confidentiality, and encryption.

1.3.7.1. How Does It Work?

To create a secure environment, an EPN model enforces a private
network with numerous restrictions. One of the core features of
an enterprise private network that businesses hope to utilize is
security. A network of this kind is set up so that no device operating
outside of the EPN can attempt to connect to the network. This
kind of model can only be accessed by registered devices according
to the network model. Access points and routers have settings
encoded in them that determine which devices are allowed access.
It is named EPN because of its integration into the business
36 COMPUTER NETWORKS

organization’s dimensions and the value it Cyberattacks can occur on networks,

adds in such a core area. A public network and sensitive information stored
and an enterprise private network are in some company data runs the
completely different from one another. An risk of being lost or obtained by
internet connection and an access point are unauthorized individuals. Even so,
necessary for a public network to function. enterprise private networks are an
excellent way to ensure that security
It is interesting to note that there risks are minimized and that the
are not many differences between public company can more easily manage
and private networks. Both regions use its critical operations since data
hardware technology and the infrastructure circulation is essential to long-term
pretty much in the same ways. The only business operations. This is a key
modern differences between them are the factor in the majority of businesses’
addition of security measures and access decision to use enterprise private
privileges. To put it another way, the terms networks, despite the setup hassles
“private” and “public” just indicate that and initial expenses that may arise.
both types of networks are understood.
Setting up a private network like EPN in • Cost-cutting measure – At the
an organization would involve several steps. end of the day, a strong return
The reason for this is that it is intended on investment requires minimal
to be highly encrypted and secure with overhead. The majority of businesses
restricted access. Thus, in order to achieve implement stringent policies to
the goal of integrating such a network, ensure their continued profitability.
organizations must concentrate on these One important method of achieving
specific areas. Enterprise private networks this is by reducing major costs, of
are becoming increasingly important to which the use of enterprise private
how businesses run and make use of their networks is a prime example.
resources, particularly when it comes to Reducing the expense of physically
sharing critical business data. Because moving resources and company
web servers are exposed to the internet data from one place to another
and can become targets of cyber security is possible by sharing crucial
breaches, it is necessary to monitor and data over the network to which
ensure their protection. Installing firewalls key stakeholders can gain access.
is also necessary to increase security. Meanwhile, a significant amount
of precious time is saved because
any type of resource or data can
1.3.7.2. Benefits of Using EPN be uploaded over the network in
a matter of minutes.
Considering the intended purpose of
enterprise private networks, it can well be • Centralization – An organization’s
said that they have a number of plausible ability to continue operating is
benefits to a business. These are discussed yet another essential component.
below: Departments within a company can
freely share resources with one
• Maximum security – It is necessary another by connecting through an
for an organization to implement enterprise private network. Time
stringent security measures. savings are made possible by it,
 COMPUTER NETWORKS AND SERVICES 37
allowing businesses to continue logic, which is typically TCP/IP, to enable
working toward meeting their daily communication. Internetworking is the
requirements gradually. process by which two networks that use
In the end, enterprise private compatible or identical communication
networks, or EPNs, continue to be cutting- protocols exchange information.
edge technology that offers genuine Routers and other internetworking
advantages to an organization. With the devices are used to implement
newest Cisco technologies, the platform internetworking as well. These are actual
provides competitive training. From QoS hardware components with the capacity to
implementation and firewalls to unified link various networks and guarantee error-
computing and unified communications, free data transfer. They are the fundamental
it covers a broad spectrum of topics and components that make it possible to work
experiences. For CCNA, CCSP, CCIP, online and form the foundation of the
and CCNP certifications, QuickStart Internet. The process of connecting several
unquestionably provides the best training. computer networks to create larger networks
To ensure that everyone understands the is known as internetworking. With the use
fundamental concepts and skills of computer of intermediary devices called gateways,
networking, a variety of courses and training various network types can be connected
programs are offered. to function as a single, sizable network.
The foundation of the current Internet was
1.3.8. Internetworks created through internetworking, which
was developed in response to a number
Internetworks join disparate networks to form of problems that arose during the early
a more expansive network. Creating a vast, development of personal computers. Many
international network is commonly referred people use various networks on a daily
to as “internetworking.” Internetworking is basis without even recognizing it. While a
the process of utilizing intermediary devices, home user might stream music to a laptop
such as routers or gateways, to connect over a wireless network, a businessperson
various networks. Internetworking uses the using a smartphone to check email uses
Internet Routing Protocol and a common a cellular network. While large wired
data communication protocol to ensure data networks are the norm in the corporate
communication between networks that are world, rural users may use dial-up to access
owned and operated by various entities. their ISP’s network. Despite their disparate
The Internet is the world’s largest collection technological backgrounds, all of these
of geographically dispersed networks, networks can connect to one another thanks
yet they are all connected by the same to internetworking.
protocol stack—TCP/IP. Working on the
internet requires that all networks that are The idea of packets—tiny, discrete
connected employ the same protocol stack data units—is essential to connecting
or communication techniques. various network types. Modern computer
networking is built on packets, which
A computer network consists of several are not limited to any particular network
computers connected to one another via technology. Alternatively, packets can be
networking hubs and switches. Every single inserted into frames—which are made for
network node or segment is configured particular network technologies—instead of
with a similar protocol or communication standalone units. This configuration makes
38 COMPUTER NETWORKS

it possible to use packets from one kind of network on another.

Packets can be transferred between these various networks by
special devices known as gateways or routers, which support
multiple networking technologies.
Internetworking gradually evolved in response to various
challenges. The first computer-to-computer connections were
established by “dumb” terminals, which had limited processing
power and were linked to powerful mainframes. Local Area Networks
(LANs) came into existence as personal computers (PCs) began
replacing terminals. Despite the numerous advantages, productivity
suffered as LANs were self-contained and unable to connect with
other LANs. Organizations with multiple locations encountered
difficulties in sharing resources like file servers, printers, and other
equipment, as well as exchanging information.
In the early 1970s, American researchers involved in the
Advanced Research Project Agency Network (ARPANET), a network
supported by the defense department, began exploring the idea of
linking their network with other early networks. This exploration
led to the development of the Transmission Control Protocol and
Internet Protocol (TCP/IP) as it became evident that existing network
protocols were not ideal for internetworking. By the late 1970s, two
Keyword additional networks had been integrated into ARPANET through
National Science TCP/IP, marking a significant milestone in the Internet’s development.
Foundation is an
independent agency Throughout the 1980s, ARPANET was used to connect an
of the United States increasing number of LANs to one another as well as new networks.
federal government
that supports ARPANET was superseded by a network constructed by the National
fundamental Science Foundation (NSF) in 1989. From there, a sizable “network of
research and networks” known as the Internet emerged when local networks were
education in all
the non-medical linked to the NSF’s network via TCP/IP and associated protocols.
fields of science and
engineering.
1.3.9. Backbone Networks (BBN)
A backbone is a key part of a network that helps nodes communicate
and provides a channel for sharing information. Backbone networks
are crucial for a network’s stability. The backbone’s role is to
bring together different parts of the network to create a unified
entity while maintaining their individuality. This allows multiple
networks to work together for specific tasks and data sharing.
Nowadays, there are many examples of backbone networks in
action. One common scenario is setting up an infrastructure that
allows networks at a company’s main office and satellite locations
to operate independently yet collaboratively. This type of network
backbone can enhance coordination efforts by enabling data sharing
between dispersed sales offices and the corporate sales office.
 COMPUTER NETWORKS AND SERVICES 39
College campuses also frequently use a among authorized users while helping
similar approach to connect local networks maintain the integrity of each network
in different buildings to a central system for within the system.
data exchange. Backbone networks typically
use various communication techniques to
establish connections and transfer data. 1.3.10. Global Area Networks
This may involve a mix of wired and wireless (GAN)
components, such as Ethernet connections An international network that links
or dial-up telephone connections for remote networks worldwide, like the Internet, is
locations with limited connectivity options. called a global area network. A network
Additionally, backbone networks often have made up of several interconnected networks
secondary connection options available for that span an extensive geographic area
flexibility. For example, even when traveling is referred to as a global area network
for business with only a hotel phone line, (GAN). The term and the Internet, which
employees may still connect to the backbone is considered a global area network, are
through alternative processes. somewhat interchangeable. Geographically
Differentiating between backbone speaking, GANs are much more extensive
networks and Internet backbones is crucial. than local area networks (LAN) and wide
Even though an internal backbone may area networks (WAN). The primary challenge
allow remote access to networks through for any GAN is moving user communications
the Internet, interconnected networks from one local coverage area to the next
within businesses remain closed unless the since a GAN is utilized to support mobile
right credentials are provided for access. communication across multiple wireless
Even if an employee can access a few LANs. A broadband GAN is the most
networks connected to the backbone, they sought-after type of GAN. The broadband
may not have access to other networks in GAN is a global satellite Internet network
the community if the backbone network’s that enables phone calls using portable
architects design the communication process terminals. These terminals offer broadband
to restrict entry. This ensures data sharing Internet access to laptop computers located
in remote areas.

Did you get it?

1. What are Home Area Networks (HAN)?
2. What distinguishes Enterprise Private Networks from other network types?
40 COMPUTER NETWORKS

FOCUS ON CAREERS
NETWORK ENGINEER
A network engineer is a professional responsible for designing, implementing, and
maintaining the intricate systems that allow computers and devices to communicate
and share resources within an organization. These professionals play a crucial role in
ensuring the reliability, security, and performance of network infrastructure. Network
engineers design network layouts and architectures, considering factors such as scalability,
redundancy, and security requirements. They are adept at configuring and managing
networking equipment such as routers, switches, firewalls, and load balancers to establish
robust connectivity and enforce security policies. Monitoring network performance and
identifying and resolving issues promptly are essential aspects of their job.
Overall, network engineers are instrumental in maintaining the backbone of an
organization’s IT infrastructure, enabling seamless communication, collaboration, and
access to resources across the network while safeguarding against potential security
risks.
A Network Engineer usually plans and manages networks to ensure they function
as intended. Network Engineers are the integral cog of an IT system. People usually
confuse Network Technicians with Network Engineers. Network Technicians usually
assist with daily IT and troubleshooting tasks while Network Engineers deal with high-
level tasks that involve designing and improvising the network.

Network Engineer Requirements

To excel in a network engineer role, candidates must meet specific job requirements,
such as:
• A bachelor’s degree in computer science (or a related field).
• Relevant certifications, such as Microsoft, Cisco, Citrix, VMware.
• Knowledge of network protocols, including TCP/IP, DNS, and DHCP.
• Experience with network hardware (routers, switches, and firewalls)
• Familiarity with network monitoring and management tools.
• Excellent problem-solving and analytical skills.
• Strong communication and teamwork abilities.

Responsibilities of a Network Engineer

Network engineers often work closely with other IT teams to integrate network services
with other systems and applications. They are also responsible for ensuring compliance
with relevant industry regulations and standards, particularly regarding data security
 COMPUTER NETWORKS AND SERVICES 41

and privacy. Continuous learning is essential in this field, as network engineers must
stay updated with the latest technologies, protocols, and best practices to adapt to
evolving network demands and emerging cyber threats. Strong analytical and problem-
solving skills, along with attention to detail, are essential for success in this role.
• Maintain and control computer networks and associated computing environments
like software for systems, software applications, equipment, and configurations.
• Running operations for disaster recovery and backups of data where possible.
• Data, applications, and hardware are protected by organizing, preparing, and
enforcing network security measures.
• Troubleshooting, diagnosing, and solving hardware, software, and other issues
affecting the network and device.
• Replacing defective hardware components of the network where possible.
• Maintain, customize, and track the device and email applications for virus
protection.
• Network performance monitoring to decide whether any changes are required.
Conferring with network users on solving issues with current networks.
• Use master consoles to control network and computer system output.
• Coordinating access to the use of computer networks.
• Plan, install, and test networking devices, hardware for computers, and devices
for operating systems.

Expect
• The work is office-based, although you may need to work across different sites,
depending on the size of the organization and its network. More frequent travel
may be involved if you work as a consultant.
• Jobs are available throughout the UK in organizations with large, sophisticated
IT systems or with consultancies providing support to clients.
• The job can be challenging, particularly when things go wrong, as companies
are dependent on their computer networks.
• Women are underrepresented in the job and the gender imbalance across the IT
industry is a recognized issue. Steps are being taken to redress the balance. See
Women in Technology and BCSWomen for more information and job vacancies.
• Systems support roles such as network engineering involve less programming,
so you should consider this before committing to a career in this area if that
is something you really enjoy.

Qualifications
You’ll usually need a degree in a subject such as:
42 COMPUTER NETWORKS

• computer science
• computer software/computer systems engineering
• computer systems and networks
• electrical/electronic engineering
• mathematics
• network security management
• physics.
It may be possible to enter this career without a degree, provided you have
significant experience. There are many Level 4 network engineering apprenticeship
opportunities and you can search for them using Find an apprenticeship. To find out
more, see Apprenticeships.
Employers will usually expect you to do further study to get professional qualifications
if you don’t already have them. For example, many colleges and private training
organizations participate in the Cisco Networking Academy program, which provides
certification at several levels for students and network professionals.

Skills
You will need:
• An up-to-date knowledge and understanding of your employer’s business and
industry needs, as well as the technical demands
• To recognize the importance of customer focus and/or of serving the needs of
the end user
• Excellent communication skills for communicating with staff who aren’t technically
trained
• The skill to take on a variety of tasks and pay attention to detail
• Analytical and problem-solving ability
• Teamwork skills and the ability to feel comfortable working with different teams,
clients, and groups of staff across an organization
• Organizational skills and the ability to prioritize your workload.

Work Experience
Relevant work experience, for example through vacation work and summer placements,
is useful as recruiters often look for evidence of skills developed through project work
and placements.
Experience in related areas such as IT support, service, and repair can be useful if
you want to move into network engineering.
 COMPUTER NETWORKS AND SERVICES 43

Being on the user end of IT systems is also helpful as this will give you an idea of
the types of problems that may arise.
Find out more about the different kinds of work experience and internships that
are available.

Employers
You can work for any organization with large, sophisticated IT systems. Typical employers
include:
• banks and building societies
• retail groups
• large government departments
• schools, hospitals and local authorities
• utility companies
• transport providers
• management consultancies.
Some organizations, including many local authorities and other public sector bodies,
outsource all their IT systems to specialists, so these specialist companies often have
vacancies for people with networking skills.
Network engineers can be employed by a large IT company to manage the firm’s
own systems or work in their contract services.
Large companies, such as Microsoft, provide an extensive range of installation and
customer support services and they recruit graduates and experienced staff into all
areas of IT support.
There are also many small consultancies that work by setting up and managing
systems for organizations that are too small to warrant full-time IT support.
Look for job vacancies at:
• CWJobs
• efinancialcareers
• Technojobs
Recruitment agencies specializing in IT, such as Hays, also have vacancies, although
these are more useful once you’ve built up some solid work experience. While large
companies usually have a structured graduate recruitment procedure, you can approach
small businesses and IT consultancies on a speculative basis, ensuring your CV stresses
your personal, as well as technical, abilities. Contracting may be an option once you’ve
gained substantial experience – visit Contractor UK for more information. Most job sites
include contractor roles as well.
44 COMPUTER NETWORKS

Other job titles used for this area of work include:

• network architect/computer network architect
• network/helpdesk support
• support/security/systems engineer
• IT/systems support engineer
• network administrator
• first/second-line support
• VoIP/Cisco engineer.

Career Prospects
Your career path will depend, to a certain extent, on the size of the organization you
work for and the scope of its IT systems. With experience, you can progress to senior
network manager and network management positions. Some network engineers choose to
broaden their careers into other IT, customer-related, or management functions. Technical
or infrastructure project management and network architecture are possibilities. Those
who start as help-desk technicians can sometimes progress to network engineer posts,
then on to senior network support, and finally network controller (mainly involved in
decision-making, staff management, and advice on future strategy). This may be the
typical route in an organization such as a large bank or a major government department.
 COMPUTER NETWORKS AND SERVICES 45

A CLOSER LOOK
Computer networking is the backbone of modern communication and information exchange,
enabling devices to connect and interact with each other across various geographic
locations. At its core, networking involves the establishment of connections between
computers, servers, routers, switches, and other devices to facilitate data transmission,
resource sharing, and collaboration. The architecture of a computer network typically
follows a hierarchical model, with multiple layers of protocols and technologies working
together to ensure seamless communication. Protocols such as TCP/IP govern how data
is transmitted and received, while devices like routers and switches manage traffic flow
within the network.
One of the fundamental concepts in networking is the distinction between local
area networks (LANs) and wide area networks (WANs). LANs connect devices within
a limited geographical area, such as a home, office, or campus, using technologies
like Ethernet or Wi-Fi. WANs, on the other hand, span larger distances and connect
multiple LANs together, often utilizing leased lines, fiber-optic cables, or satellite links.
The internet itself is the most extensive WAN, providing global connectivity to billions
of devices worldwide.
Security is a critical aspect of computer networking, as networks are susceptible to
various threats, including unauthorized access, data breaches, and malicious attacks.
Network security measures such as firewalls, intrusion detection systems, and encryption
protocols help safeguard sensitive information and prevent unauthorized access to
network resources.
Advancements in networking technology are constantly pushing innovation and
shaping the digital landscape. Emerging trends like 5G connectivity, edge computing,
software-defined networking (SDN), and the Internet of Things (IoT) are changing how
networks are designed, deployed, and managed. These developments bring faster data
speeds, lower latency, increased flexibility, and scalability, creating new opportunities
for applications like autonomous vehicles, augmented reality, and smart cities.
46 COMPUTER NETWORKS

SUMMARY
• A computer network is a group of linked computers. Nodes are computers
connected to a network, either wirelessly using radio waves or through cabling
like an Ethernet cable.
• Wide Area Networks (WANs) connect computers and local networks to larger
networks across extensive geographical regions, like continents.
• A computer network is a group of computers connected by communication
channels to share resources and communicate with users.
• The fundamental idea behind OSI is that communication between two endpoints
in a network can be separated into seven different groups, or layers, of related
functions. Every user or program that communicates uses a device capable of
providing those seven levels of functionality.
• Specialized routing protocols are designed for internet network routers. Each
device in a ring topology has a unique point-to-point connection with only the
other two devices on either side.
• Every device in a star topology has a dedicated point-to-point link that connects
it to the hub, or central controller.
• Hostname is a special device name that is assigned to each device in the
network. In the Administrator Mode command prompt, type “hostname” and
hit “Enter” to see your machine’s hostname.

MULTIPLE CHOICE QUESTIONS

1. Combination of two or more networks are called
a. Internetwork
b. WAN
c. MAN
d. LAN
2. Which topology covers security, robust and eliminating traffic factor?
a. Mesh
b. Ring
c. Star
d. Bus
3. National Internet Service Provider (ISP) networks are connected to one another
by private switching stations called
a. Network Access Points
b. Peering Points
c. National ISP
c. Regional ISP
 COMPUTER NETWORKS AND SERVICES 47
4. Multipoint topology is
a. Bus
b. Star
c. Mesh
d. Ring
5. A communication path way that transfers data from one point to another is called
a. Link
b. Node
c. Medium
d. Topology
6. Which of the following devices operates at the Data Link layer of the OSI model?
a. Router
b. Hub
c. Switch
d. Bridge
7. What protocol is used for transferring files over the Internet?
a. FTP
b. HTTP
c. SMTP
d. TCP
8. Which networking device works at the Application layer of the OSI model?
a. Router
b. Switch
c. Hub
d. Proxy server
9. In TCP/IP, what is the maximum number of addresses in the IPv4 address space?
a. 256
b. 232
c. 264
d. 2128
10. What protocol is used to translate domain names to IP addresses?
a. DNS
b. DHCP
c. ARP
d. ICMP
48 COMPUTER NETWORKS

REVIEW QUESTIONS
1. What is the Open Systems Interconnection (OSI) model, and what are its different
layers?
2. What is network topology, and what are the common types of network topologies?
3. How does email management contribute to efficient communication within
organizations?
4. What are the key responsibilities of system administrators in managing computer
systems and networks?
5. What are Personal Area Networks (PANs), and what are some examples of
devices commonly used in PANs?

Answer to Multiple Choice Questions

1. (a) 2. (a) 3. (b) 4. (a) 5. (a)
6. (d) 7. (a) 8. (d) 9. (b) 10. (a)

REFERENCES
1. Daniel, B., (2008). The Salsa20 family of stream ciphers; Chapter. In: Matthew, R.,
& Olivier, B., (eds.), New Stream Cipher Designs. Springer-Verlag.
2. Guido, A., Isaac, K., & McKeown, N., (2004). Sizing router buffers. ACM SIGCOMM
Computer Communication Review.
3. Joan, D., & Vincent, R., (2002). The Design of Rijndael: AES – The Advanced
Encryption Standard. Springer-Verlag.
4. Mills, D. L., & Kamp, P. H., (2000). The nanokernel. Proc. Precision Time and Time
Interval (PTTI) Applications and Planning Meeting (423–430). Reston VA.
5. Mogul, J., Mills, D., Brittenson, J., Stone, J., & Windl, U., (2000). Pulse-Per-Second
API for Unix-Like Operating Systems, Version 1 (p. 31). Request for Comments
RFC-2783, Internet Engineering Task Force.
6. Neal, C., Yuchung, C., Stephen, G. C., Soheil, H. Y., & Van, J., (2016). BBR congestion-
based congestion control. ACM Queue, 14(5).
7. Stefania, C., Bruce, D., Arjen, K. L., Walter, L., Peter, M., Brian, M., Herman Te, R.,
et al., (2000). Factorization of a 512-bit RSA modulus. Advances in Cryptology —
EUROCRYPT 2000, Lecture Notes in Computer Science (Vol. 1807). Springer-Verlag.
8. Vladimir, B., Suman, B., Marco, G., & Sangho, O., (2008). Wireless device identification
with radiometric signatures. Proceedings of the 14th ACM International Conference
on Mobile Computing and Networking (MobiCom ‘08).
CHAPTER
Network
2 Model and
Standardization

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Prepare for the OSI model

• Know about the multiplexing and switching
• Discuss on transmission control protocol (TCP)
• Learn about the user datagram protocol (UDP)
50 COMPUTER NETWORKS

INTRODUCTORY EXAMPLE

THE OSI MODEL FOR CLOUD COMPUTING AND VIRTUALIZATION

Cloud computing is now crucial for online businesses as it helps them store, access, and
process data efficiently at lower costs. The OSI model enables seamless operation and
interoperability between different systems. All layers of the OSI model closely align with
cloud computing, which is essentially the delivery of on-demand computing resources
over the Internet. OSI provides a framework for encompassing the various components
involved in cloud computing. At the lower layers of OSI, we find virtualization technology.
Virtualization creates virtual instances of operating systems and storage within a physical
infrastructure, allowing for efficient resource utilization, scalability, and flexibility in
the cloud environment. There are two main layers of the OSI model that enable cloud
computing: the Application Layer and the Network Layer. The Network Layer routes and
transmits data across networks to deliver them to client devices, while the Application
Layer defines the standards for clients to access cloud-based applications.
INTRODUCTION
A network’s core is its networking software. It helps admins deploy, maintain, and monitor
networks. Traditional networks use specialized hardware like switches and routers with
integrated networking software. Software-defined networking (SDN) separates software
from hardware, making it easier to innovate and adapt quickly. Network software allows
admins to access network operations, unlike software applications for end users. Its
main role is to facilitate users’ access to network resources seamlessly.
• The basic functionality of network software includes:
• User management: Enables administrators to add or remove users from the
network.
File management: Allows administrators to define the location of data storage and
user access to that data.
Network standards establish the guidelines for communication between devices
connected to a network, ensuring the interoperability of networking technologies. In order
to reduce the possibility of incompatibility, networking standards are in place to help
guarantee that products from various vendors can operate together in a network. Our
dictionary of networking standards provides a glossary of key terms you should know.
If the various network standards in use today were published in print, they would fill
volumes upon volumes of text. They range from wireless networking standards like 802.11a
and Gigabit Wi-Fi to wired networking standards like Ethernet and Gigabit Ethernet.
The standards themselves specify the functions for which these network standards are
intended. For instance, you need to use one standard to talk over the phone network
and another to communicate over the Internet. Nearly everything that we encounter
in a day is determined by standards. Both your TV’s television signal and the cover on
your “to-go” coffee cup adhere to standards. While some standards, like the television
signal, are unregulated, others are, such as the new High-Definition Television (HDTV)
standard developed in part by the Advanced Television System Committee (ATSC) or
the National Television System Committee (NTSC).

NETWORK MODEL AND STANDARDIZATION 51

52 COMPUTER NETWORKS

2.1. THE OSI MODEL

→Learning Objectives

• Understand the purpose of application layer.

• Learn the responsibilities of the presentation layer.
• Explore the management of session layer.
• Comprehend the mechanisms used in the transport layer.
• Study the useful concepts in the network layer.
• Understand the framing detection in the Data Link Layer.
• Familiarize signaling in the Physical Layer.

The ISO/OSI model, which stands for The challenge arises when a vendor’s
International Standards Organization/Open products cannot easily integrate with those of
System Interconnection, is a commonly used other vendors, impacting their marketability.
reference model for network communication This, in turn, presents a significant issue
between two end users. Understanding for vendors. The ISO model provides a
the network’s operation can be helpful for networking architecture for implementing
addressing future issues. The significance of protocols layer by layer. While each layer
the OSI model cannot be overstated. Most functions independently, it also supports
networking suppliers and users recognize the layer above it and is supported by the
the importance of network computing layer below it. The layers are divided into
products adhering to and fully supporting two groups. When a message is sent to
the networking standards established by or from a user, the upper layers come into
this model. play. On the other hand, when a message
passes through the host computer, the lower
Connecting a vendor’s products to those layers are involved. Messages destined for
of other vendors becomes relatively easy the device target the higher layers, while
when they comply with the standards set those meant for another host do not get
by the ISO model. Conversely, the more a forwarded to the upper layers.
vendor strays from these guidelines, the
more challenging it becomes to integrate Layers of OSI model are as follows:
their offerings with those of other vendors. • Application layer
Deviating from the communication standards
fostered by the model would make software • Presentation layer
development efforts significantly more • Session layer
difficult for a vendor. They would need to • Transport layer
develop all the necessary software from
• Network layer
scratch instead of building on existing work
done by other vendors. • Data link layer
• Physical layer
 NETWORK MODEL AND STANDARDIZATION 53
2.1.1. Application Layer
The application layer provides users, whether software or human, with network access.
It supports services like email, remote file access, transfer, shared database management,
and other distributed information services, offering user interfaces for them.

Figure 2.1. The request context of application layer (Source: Cloudflare, (https://cf-assets.www.
cloudflare.com/slt3lc6tev37/2rcDKpr4WLqoyAZ7GDKkyJ/7cab96402de7ac5465b86e617da3da4e/
osi_model_application_layer_7.png).

Specific services provided by the application provides distributed database

layer include: sources and access to global
information about various objects
• Network Virtual Terminal: A network and services.
virtual terminal is a software
version of a physical terminal that
allows a user to log on to a remote 2.1.2. Presentation Layer
host. To do so, the application Data received is converted into a format that
creates a software emulation of a the intended application can understand at
terminal at the remote host. The the presentation layer. Think of the work at
user’s computer communicates this layer as a translation task. For example,
with the software terminal, which data is often encrypted at the presentation
then communicates with the host, layer before being sent to other layers for
and vice versa. The remote host transmission. Once received, the data is
believes it is communicating with decrypted and sent in the expected format
one of its own terminals and allows to the application.
the user to log on.
The presentation layer’s specific
• File Transfer, Access, and
responsibilities include translation and
Management ( F TA M ) : This
encryption.
application allows a user to access
files on a remote computer (to make • Translation: Translation involves
changes or read data), retrieve ensuring interoperability between
files from a remote computer, and different encoding techniques used
manage or control files on a remote by various computers.
computer. • Encryption: Encryption is essential
• Mail Services: This application for maintaining privacy when
provides the basis for email transmitting sensitive information.
forwarding and storage. It involves transforming the original
• Directory Services: This application information into another form for
54 COMPUTER NETWORKS

sending over the network, with decryption reversing this

process to restore the message to its original form.
• Compression: Data compression reduces the number of bits
to be transmitted. Data compression becomes particularly
Keyword important in the transmission of multimedia such as text,
A remote audio, and video.
computer is a
computer that you
can connect to from
another location
and control as if
you were physically
present. It’s also
known as a remote
desktop.

Figure 2.2. Presentation layer (Source: Thenationaltv, https://

f002.backblazeb2.com/b2api/v1/b2_download_file_by_id?fileId=4_
za8a2358db1d7f91b68b30916_f102f662b1ec69f3a_d20190909_m073653_
c002_v0001114_t0001).

2.1.3. Session Layer

Establishing, maintaining, and terminating connections take place at
the session layer. This usually refers to network-based application
requests for data. As the actual data flow is managed by the
transport layer, the session layer serves as an announcer, informing
the programs and applications sending and receiving data when
their requests are fulfilled. The session layer synchronizes data
transmission, to put it technically.
Specific responsibilities of the session layer include:
• Dialog Control: The session layer allows two systems to
enter into a dialog. It allows the communication between
two processes to take place either in half-duplex (one way
at a time) or full-duplex (two ways at a time). For example,
the dialog between a terminal connected to a mainframe
can be half-duplex.
 NETWORK MODEL AND STANDARDIZATION 55
• Synchronization: The session layer allows a process to add
checkpoints into a stream of data.

Figure 2.3. Session layer (Source: Open4tech, https://open4tech.com/wp- Remember

content/uploads/2019/09/session-layer.png). Transport layer
services are
conveyed to an
2.1.4. Transport Layer application via
a programming
Transmission of data over the network is handled by the transport interface to the
layer. The data is conceptualized at this level more as a conversation transport layer
than as individual packets. This is achieved by using protocols, protocols.
which are referred to as “rules of communication.” The protocols
monitor the entire transmission of numerous packets, looking for
errors in the conversation, acknowledging successful transmissions,
and requesting retransmission in the event that errors are found.
Specific responsibilities of the transport layer include:
• Service-Point Addressing: Multitasking is a common feature
of computers. Because of this, delivery from one computer to
another or from one computer’s running program to another
computer’s running program is referred to as source-to-
destination delivery. Hence, a sort of address known as a
service-point address (or port address) needs to be included
in the transport layer header. While the transport layer
delivers the complete message to the appropriate process
on that computer, the network layer delivers each packet
to the right machine.
• Segmentation and Reassembly: A message gets split into
transmittable chunks, each with a sequence number
attached. These numbers help the transport layer identify
and replace lost packets during transmission, as well as
correctly piece together the message upon reaching the
destination.
• Connection Control: There are two types of transport
56 COMPUTER NETWORKS

layers: connection-oriented and connectionless. With a connectionless transport

layer, each segment is sent to the destination machine’s transport layer as an
independent packet. In contrast, a connection-oriented transport layer establishes
a connection with the destination machine’s transport layer before sending the
packets. The connection is closed once all data is transmitted.
• Flow Control: Flow control, similar to the data link layer, is managed by the
transport layer. However, at this level, flow control is implemented end-to-end
rather than over a single link.
• Error Control: The transport layer shares the responsibility for error control with
the data link layer. Error control at this layer occurs end-to-end, ensuring that
the entire message reaches the receiving transport layer without any errors
(loss, duplication, or damage). Retransmission is commonly used to rectify errors.

Figure 2.4. Transport Layer in OSI Networking System (Source: Fiberbit Technology Co.Ltd., http://
fiberbit.com.tw/wp-content/uploads/2013/06/transport-layer-in-osi-networking-model-system.jpg).

2.1.5. Network Layer

The network layer is responsible for sending a packet from its source to its destination,
possibly over several networks. The network layer ensures that every packet reaches
its destination from its point of origin, while the data link layer handles packet delivery
between two systems connected by the same network (links). Network layers are
typically unnecessary when two systems are linked to the same link. However, source-
to-destination delivery is often achieved by the network layer if the two systems are
connected to separate networks (links) with connecting devices between the networks
(links).
Specific responsibilities of the network layer include:
• Logical Addressing: The data link layer’s physical addressing scheme resolves
the local addressing issue. To differentiate the source and destination systems,
a packet requires a different addressing system when it crosses the network
boundary. In addition to the logical addresses of the sender and recipient, the
network layer adds a header to the packet originating from the upper layer.
• Routing: Devices that link separate networks or links to create an internetwork,
also known as a network of networks, or a large network, route packets to their
 NETWORK MODEL AND STANDARDIZATION 57
intended destination. These devices are known as routers or gateways. This is
one of the functions performed by the network layer.

Figure 2.5. Network layer process (Source: Geeksforgeeks, (https://media.geeksforgeeks.org/wp-

content/uploads/20230419121402/Forwarding-2-(2).png).

2.1.6. Data Link Layer

Information gets transformed at the data link layer into coherent “packets” and frames
that are sent to higher layers. Essentially, the data link layer changes information from
the upper layers into raw data that can be sent over the physical layer and unpacks
raw data that comes from the physical layer. Another task of the data link layer is to
correct and compensate for errors that occur in the physical layer.

Did you know?

In some networks, such as IEEE 802 local area networks, the data link layer is described in more detail with
media access control (MAC) and logical link control (LLC) sublayers. This means that the IEEE 802.2 LLC
protocol can be used with all of the IEEE 802 MAC layers, such as Ethernet, Token Ring, IEEE 802.11, etc., as
well as with some non-802 MAC layers such as FDDI..

Figure 2.6. Framing of data link layer (Source: Open4Tech, (https://open4tech.com/wp-content/

uploads/2019/09/data-link-layer.png).
58 COMPUTER NETWORKS

Specific responsibilities of the data link layer include:

• Framing: The data link layer divides the stream of bits
received from the network layer into manageable data
units called frames.
• Physical Addressing: If frames are to be distributed to
different systems on the network, the data link layer adds
a header to the frame to define the physical address of the
sender (source address) and/or receiver (destination address)
of the frame. If the frame is intended for a system outside
the sender’s network, the receiver address is the address
of the device that connects one network to the next.
• Flow Control: If the rate at which the data is absorbed by
the receiver is less than the rate produced by the sender,
the data link layer imposes a flow control mechanism to
prevent overwhelming the receiver.
• Error Control: The data link layer adds reliability to
the physical layer by adding mechanisms to detect and
retransmit damaged or lost frames. It also uses a mechanism
to prevent duplication of frames. Error control is normally
achieved through a trailer added to the end of the frame.
• Access control: When two or more devices are connected
to the same link, data link layer protocols are necessary
Keyword to determine which device has control over the link at any
Fiber optics is given time.
the technology
used by internet
services such as
Verizon Fios home
2.1.7. Physical Layer
internet to transmit The physical layer of a network includes cables, fibers, cards,
information as
pulses of light switches, and other components. This layer changes digital data
through strands of into signals for transmission over wires. These signals can be
fiber made of glass electrical or non-electrical like optics or other encoded pulses. The
or plastic over long
distances. physical layer’s role in networking is to enable data transmission.
Troubleshooting the physical layer is likely easier due to the need for
hardware connections, but building or repairing it can be challenging.

Figure 2.7. Physical layer (Source: simplilearn, https://www.simplilearn.

com/ice9/free_resources_article_thumb/Working_And_Implementation_Of_
Physical_Layer_In_The_OSI_Model_2.png).
 NETWORK MODEL AND STANDARDIZATION 59
The physical layer is concerned with:
• Physical Characteristics of Interfaces and Media: The physical layer defines
the characteristics of the interface between the devices and the transmission
medium. It also defines the type of transmission medium.
• Representation of Bits: The physical layer data consists of a stream of bits (a
sequence of 0’s and 1’s) without any interpretation. The physical layer defines
the type of encoding (how 0’s and 1’s are changed to signals).
• Data Rate: The transmission rate, the number of bits sent each second, is also
defined by the physical layer. In other words, the physical layer defines the
duration of a bit, which is how long it lasts.
• Synchronization of Bits: The sender and receiver must be synchronized at the bit
level. In other words, the sender and the receiver clocks must be synchronized.
• Line Configuration: The physical layer is concerned with the connection of devices
to the medium. In a point-to-point configuration, two devices are connected
together through a dedicated link. In a multipoint configuration, a link is shared
between several devices.
• Physical Topology: The way that devices are connected to form a network is
specified by the physical topology. Devices can be linked together using one
of four topologies: bus (every device on a common link), ring (every device
connected to the next, forming a ring), mesh (every device connected to every
other device), or star (devices connected through a central device).
• Transmission Mode: The physical layer also specifies whether two devices will
transmit in a simplex, half-duplex, or full-duplex fashion. One device can only send
while the other can only receive in simplex mode. One-way communication is used
in the simplex mode. Two devices can send and receive simultaneously when
using the half-duplex mode. Two devices can send and receive simultaneously
when they are in full-duplex (or just duplex) mode.

Did you get it?

1. When to use transport layer in OSI model?
2. Where do you use data link layer in OSI model?
60 COMPUTER NETWORKS

2.2. MULTIPLEXING AND

SWITCHING
→Learning Objectives

• Examine the various types of multiplexing

• Discuss on the techniques of switching
• Know about the digital multiplexers
• Learn about the digital demultiplexers
• Execute the multiplexers as PLDS

Multiplexing is the process of combining several signals (analog or digital) for transmission
over a single line or medium. Multiple low-speed signals are combined for transmission
over a single high-speed connection in a common form of multiplexing. A multiplexer
(MUX), which combines N input lines to produce one output line, is used to multiplex
data (many to one).

Figure 2.8. Multiplexing (Source: Geeksforgeeks, https://media.geeksforgeeks.org/wp-content/

cdn-uploads/CN_Multiplexing-1.jpg).
 NETWORK MODEL AND STANDARDIZATION 61
As a result, a multiplexer (MUX) has one output and multiple
inputs. A device known as a demultiplexer (DEMUX) is used at
the receiving end to split the signal into its component signals.
DEMUX therefore has multiple outputs and one input. Four input
lines are fed into the multiplexer, which outputs one output line.
This single line combines and transmits the signal from four distinct
devices. A demultiplexer at the receiving end splits this signal from
a single line into its original signals, which are then sent to the
four separate receivers.

2.2.1. Types of Multiplexing

Types of multiplexing are as follows:
• Frequency-division multiplexing (FDM): An analog
multiplexing technique called frequency-division multiplexing
(FDM) combines analog signals. When the combined
bandwidths of the signals to be transmitted are less than
the bandwidth of a link (measured in hertz), FDM is used.
Different carrier frequencies are modulated by the signals
produced by each sending device in FDM. After that, the
modulated signals are combined into a single composite
signal that the link is able to carry. Enough bandwidth
separates the carrier frequencies to allow for the modulated
signal. The different signals travel through these channels,
which are these bandwidth ranges. Strips of unused Keyword
bandwidth can be used to divide channels, i.e., guard bands,
Bandwidth is how
to prevent signals from crossing over. Furthermore, carrier much information
frequencies shouldn’t obstruct the original data frequencies. you receive every
second, while
• Wavelength-Division Multiplexing (WDM): It is an analog speed is how fast
method of combining optical signals through multiplexing. that information
WDM is designed to leverage the high data rate capacity is received or
downloaded.
of fiber-optic cables. The data rate of metallic transmission
cables is lower than that of optical fiber. When a fiber-optic
cable is used for a single line, its available bandwidth is
wasted. Multiplexing allows us to combine multiple lines into
a single one. Although optical signals sent through fiber-optic
channels are used for the multiplexing and demultiplexing
processes, WDM and FDM are conceptually identical. The
concept remains the same: combining various signals with
different frequencies. These frequencies are extremely high,
which makes a difference.
• Time Division Multiplexing (TDM): This method of digital
multiplexing converts multiple low-rate channels into a single
high-rate channel. Unlike FDM, which shares a portion of the
bandwidth, TDM is a digital process that enables multiple
62 COMPUTER NETWORKS

connections to share the high bandwidth of a link. Instead of

sharing bandwidth, time is shared. Each connection occupies
a specific time slot in the link. By combining digital data from
various sources, a time-shared link is created. It is worth
noting that analog data can still be generated by the sources;
it just needs to be sampled, converted to digital data, and
then multiplexed using TDM.

2.2.1.1. Types of TDM

Types of TDM are as follows:
• Synchronous TDM: A frame is created by gathering a
round of data units from every input connection. A frame
is split up into N time slots if there are N connections,
one for each unit and each input line. Each slot has a Tin
Remember duration, each frame has a T duration, and the input unit
Time-division has a T duration. To ensure data flow, the output link’s
multiplexing is data rate needs to be N times higher than the connection’s
used primarily for data rate. Frames are used to organize time slots. A frame
digital signals,
but may be is made up of one full cycle of time slots, one for every
applied in analog sending device. Each frame in a system with N input lines
multiplexing in has N slots, and each slot is designated to carry data from
which two or a particular input line.
more signals
or bit streams • TDM interleaving: Two fast-rotating switches, one on the
are transferred multiplexing side and one on the de-multiplexing side, can
appearing be seen. The switches spin simultaneously and in opposite
simultaneously
as sub-
directions at the same speed. As the switch opens in front
channels in one of a connection on the multiplexing side, that connection
communication has the chance to send a unit onto the path. Interleaving
channel, but are is the term for this procedure. By opening the switch in
physically taking front of a connection, the de-multiplexing side allows that
turns on the
channel. connection to receive a unit from the path.
• Statistical Time-Division Multiplexing: A reserved slot
is assigned to each input in the output frame during
synchronous TDM. If there are input lines with blank data,
this could be inefficient. To increase bandwidth efficiency,
statistical time-division multiplexing dynamically assigns
slots. An input line receives a slot in the output frame only
if it has a slot’s worth of data to transmit. The number of
slots in each frame during statistical multiplexing is less
than the total number of input lines. Round robin, the
multiplexer checks each input line; if a line contains data
to send, it is assigned a slot; if not, it skips the line and
checks the one after it.
 NETWORK MODEL AND STANDARDIZATION 63
2.2.2. Switching advantage is that, because each
packet is brief, the communication
A collection of connected nodes, referred to
links between the nodes are only
as switches, make up a switched network.
used to send one message at a time
Switches are devices that can establish
during packet transmission. Longer
temporary connections between two or more
messages need to be sent in a series
connected devices. Some of these nodes are
of packets, but they don’t need a
linked to the end systems (computers or
dedicated link to be sent in between
phones, for instance) in a switched network.
each packet. This allows for a much
Others are only used in routing. Types of
more equitable distribution of the
switching of network are as follows:
resources across all of the links. It
• Circuit Switching: Physical links is implied that packets from other
connect a group of switches in messages may be sent between the
a circuit-switched network. A packets of the message being sent
dedicated path consisting of one or from A to Ddot.
more links connects two stations. On • “Pipelining” is a further benefit
each link, though, each connection of packet switching. When packet
only uses one dedicated channel. 1 is sent from C to D, packet 2
FDM or TDM is typically used to is sent from B to C, packet 3 is
divide each link into N channels. sent from A to B, and so on. At
Electrical switching occurs at the the same time, packet 1 is sent
physical layer. from C to D. This concurrent use
• Packet Switching: A message is of the communication links results
split up into multiple segments in increased efficiency. Despite the
or packets when using packet fact that a header is included in
switching. Every packet is handled each packet rather than in each
as an independent communication. message, the overall latency for
Then, a continuous sequence of transmission across a packet
these packets is sent to each network may be significantly lower
station. It is possible for packets than for message switching.
from different messages to be – Message Switching: All
transmitted together or for the communications sent to any
same message to be sent over computer system connected to
multiple lines. A packet’s source the network are received by
and destination addresses, as well the central switching station
as its packet address, are listed in a message switching
in its header. When the packets network. A message is then
arrive at their destination, they put stored in its buffer memory.
themselves back together to form Every message includes its
the original message. intended recipient’s address.
The packet switching common network Messages are forwarded to
is made up of stations, nodes, and their addresses as soon as
transmitted parts. The benefits of packet lines become available. The
switching are as follows: store and forward method is
another name for message
• The first, and most significant, switching.
64 COMPUTER NETWORKS

2.2.3. Digital Multiplexers

Multiplexer is short for multiplexing. A circuit known as a multiplexer is used to
choose and direct any one of the many input signals toward an output signal. A single
pole multiposition switch is a basic illustration of a non-electronic multiplexer circuit.
Numerous electronic circuits make extensive use of multiposition switches. On the other
hand, circuits with high operating speeds need an automatic multiplexer selection. This
is not a task that a mechanical switch can complete effectively. As a result, electronic
components are used in the construction of the multiplexers that perform high-speed
switching. Analog and digital data are handled by multiplexers. Relays and transistor
switches are the building blocks of multiplexers used in analog applications. Standard
logic gates are used in their construction for digital applications. A circuit with multiple
inputs but only one output is known as a digital multiplexer, and it is utilized in digital
applications. Control signals allow any input to be directed toward any output. The
2-to-1, 4-to-1, 8-to-1, and 16-to-1 multiplexers are among the few types of multiplexers.

2.2.3.1. Understanding 4-to-1 Multiplexer:

There are 4 input bits, 2 control bits, and 1 output bit in a 4-to-1 multiplexer. The bits D0,
D1, D2, and D3 are the four inputs. Output Y receives only one of these transmissions.
The value of AB, the control input, determines the output. Which input data bit is
sent to the output depends on the control input. For example, as Figure 2.9 illustrates.
The upper AND gate is active and the other AND gates are inactive when AB = 00.
Consequently, data bit D0 is sent to the output, resulting in Y = D0.

Figure 2.9. 4-to-1 multiplexer (Source: EEWeb, https://www.eeweb.com/wp-content/uploads/

articles-articles-4-to-1-multiplexer-circuit-diagram-1387783580.jpg).
 NETWORK MODEL AND STANDARDIZATION 65
If the control input is changed to AB = 11, all gates are disabled
except the bottom AND gate. In this case, D3 is transmitted to the
output and Y = D3.
• An example of a 4-to-1 multiplexer is IC 74153 in which
the output is the same as the input.
• Another example of a 4-to-1 multiplexer is 45352 in which
the output is the complement of the input.
• An example of a 16-to-1 line multiplexer is IC74150.

2.2.3.2. Applications of Multiplexers

Multiplexers are used in various fields where multiple data need
to be transmitted using a single line. There are some applications
of multiplexers:
• Communication System: Transmission systems, relay Keyword
and tributary stations, and communication networks are Tributary
examples of systems that facilitate communication. With station means an
a multiplexer, a communication system’s efficiency can be aeronautical fixed
station that may
significantly increased. Multiplexers enable the simultaneous route, receive, or
transmission of several types of data, including audio and transmit messages
video, over a single transmission line. or digital data
but does not
• Telephone Network: Using multiplexers, several audio relay except for
signals are combined onto a single line for transmission in the purpose of
serving similar
a telephone network. This allows for the isolation of several stations connected
audio signals until the desired audio signals are eventually through it to a
received by the intended parties. communication
center.
• Computer Memory: A large amount of memory can be added
to a computer by using multiplexers, which also reduce the
number of copper lines needed to connect the memory to
other components of the computer circuit.
• Transmission from the Computer System of a Satellite: A
multiplexer can be used for the transmission of data signals
from the computer system of a satellite or spacecraft to the
ground system using the GPS (Global Positioning System)
satellites.

2.2.3.2. Chaining Multiplexers

Up to four MPEG encoders can be multiplexed and controlled by the
MIDI MUX. Multiple multiplexers must be chained together if more
than four services are to be encoded on a single transport stream.
66 COMPUTER NETWORKS

2.2.3.3. How to Connect the Multiplexers:

The TS input of the subsequent multiplexer must be connected to the TS output port
of the multiplexer before it can be chained. It makes no difference which TS input port
is used for chaining because all of the input ports on the Mini MUX and MIDI MUX are
equal. The output device should be connected to the TS output of the final multiplexer
in the chain. The transport stream clock should go with the data so TS input port and
TS output port should be configured as follows for chaining:
• TS Output: Clock Direction: generate
• TS Output: Frame size: 188 Byte
• TS Output: Clock rate
• TS Input Mode: External click
The total clock rate of all TS sources linked to the multiplexer plus an extra 10%
as a safety precaution should equal the transport stream clock rate. Assume a Midi
Mux has four linked encoders, each with a 4,000 Kbit/s data rate set. This results in
a total data rate for this Midi Mux’s TS output of 16,000 Kbit/s. The TS Output Clock
rate should be set to 18,000 Kbit/s with the safety margin. After receiving this stream
at one of its TS inputs, the subsequent Midi Mux in the chain adds three more services
from encoders linked to the remaining three TS inputs, each with 4,000 Kbit/s. This
means that the output of this Midi Mux is seven times 4,000 Kbit/s. One sets the TS
output to 32,000 Kbit/s with the safety margin.

2.2.4. Digital Demultiplexers

When you have one too many, it’s demultiplexed. A circuit with one input and multiple
outputs is known as a demultiplexer. Any input can be directed to any output by using
a control signal. There are various types of demultiplexers, such as 1-to-2, 1-to-4, 1-to-
8, and 1-to-16. The general idea of a demultiplexer with one input signal, m control
signals, and N output signals is illustrated in Figure 2.10.

Figure 2.10. Digital demultiplexers (Source: IPUNOTES, https://i0.wp.com/www.electronicshub.

org/wp-content/uploads/2013/12/Demultiplexer-Pin-Diagram.png).
 NETWORK MODEL AND STANDARDIZATION 67
Any multiplexer input line to any
2.2.5. Multiplexers as PLDs multiplexer output line can only have two
Typically, one or more interconnect arrays entry points with conventional multiplexers.
used by programmable logic devices (PLDs) Put differently, a previous art multiplexer
are programmed using an array of memory had two output lines that could be
cells (e.g., EEPROM, flash EPROM, flash connected to each input line. Routing
EEPROM, or EPROM cells) to create the signals through the multiplexer becomes
different PLD interconnections that are very challenging if a particular user design
necessary for a particular design. Because calls for multiple input signals to be routed
PLD complexity can be increased on a regular into a function block (which includes AND
basis due to technological advancements, array 102, multiplexer 101, and macro cells
increasing the size of the programmable 103). One must switch different multiplexer
interconnect array is also necessary to reach input signals to different lines if routing
the desired PLD complexity. Regretfully, through the multiplexer is not possible.
the programmable elements of the Usually, this is done with the help of the
programmable interconnect array consume programming software used to configure
a large portion of the die size and power. the PLD.
Furthermore, the programmable elements
used in the programmable interconnect In such a case, the input lines are
array cause capacitive loading, which accessed via different multiplexer ports.
significantly reduces speed. As a result, Unfortunately, this approach is undesirable
a larger programmable interconnect array because:
uses more power and operates at a slower
• If an input signal is an external
pace. Therefore, the programmable elements
signal that is received via a specific
themselves become a more important and
PLD pin, it may not be possible to
limiting factor as larger PLDs are designed.
place the signal on the desired pin;
Multiplexer 101’s requirement for a • If an input signal is received from
discrete control signal for each multiplexer a macro cell within the PLD, the
connection is one issue with this PLD. For signal must be placed in a different
PLD designs that require a small number macro cell, thereby exacerbating
of multiplexers, this arrangement is fine; the placement problem; and
however, PLD designs that require a large • Moving an input signal to a different
number of multiplexers also require a larger input line of the multiplexer implies
number of control signals. Multiplexers that the signal is moved globally,
are commonly arranged in arrays with 450 i.e., for all multiplexers associated
multiplexer connections, which consist of with all other function blocks. For
126 input lines × 3 connections per input large PLDs, there are many function
line (including input pad and feedback blocks, thereby increasing the
connections) + 36 ground connections + probability that a particular routing
36 UIM connections. Up to six of these for one multiplexer adversely
multiplexers can be found in a typical PLD, affects the routing of the other
necessitating 2700 multiplexer connections multiplexers to which the signal
and 2700 control signals. These control is propagated.
signals are unnecessarily complicated to
route through a PLD. Consequently, in order to enable the
reliable implementation of complex user
68 COMPUTER NETWORKS

designs without significantly increasing the PLD’s size or power requirements or adding
needless complexity, a multiplexer that offers effective and flexible control and routing
options is needed.

Did you get it?

1. What is the function of digital multiplexers?
2. How can multiplexers be implemented as PLDs?
 NETWORK MODEL AND STANDARDIZATION 69

2.3. TRANSMISSION
CONTROL PROTOCOL
(TCP)
→Learning Objectives

• Find the TCP header format

• Clarify the connection establishment and termination
• Examine the sliding window and flow control
• Discuss on congestion control

The Internet connection suite consists of two protocols, one of which is the Transmission
Control Protocol. The basis of almost all Internet traffic is formed by the combination
of this protocol and the Internet Protocol. Because these protocols are practically
indispensable in their usage and depend entirely on one another to function properly,
the Internet connection suite is commonly referred to as TCP/IP (Transmission Control
Protocol/Internet Protocol).
The Internet Protocol is in charge of routing and transmission, while the Transmission
Control Protocol disassembles and reassembles data. Small data bursts known as packets
make up the majority of Internet traffic. Along with some optional extra information,
these packets include details about the data’s origin and destination. The Internet
Protocol sends the packets over the Internet, while the Transmission Control Protocol
creates and reassembles them.
Consider moving a document from a distant computer to a local one to demonstrate
this procedure. The Internet protocol decides how the document needs to be moved
between the computers after the computer receives the move command. Next, the
transmission control protocol is requested by the Internet protocol to divide the document
in a particular manner. The document is then divided into packets by TCP, which are
then sent out via the Internet protocol. TCP starts reassembling the document packets
as soon as they reach the local computer. It first makes sure that it has all of them
and that they are in good condition. If not, it requests the necessary parts and waits
for them to be delivered. Once it has all the parts, it assembles the document by
70 COMPUTER NETWORKS

rearranging them according to the correct order. The computer is

then notified that the file has arrived by the Internet protocol. The
great majority of Internet usage consists of these procedures. Every
interaction in programs like file transfer systems, email clients,
and web browsers uses this fundamental back-and-forth process.
Information can be sent between unrelated programs and systems
without conversion thanks to this unified system. Given that this
is the fundamental building block of the Internet, there isn’t much
variation in the protocol.
This allows for backward compatibility between systems across
the Internet. Older machines might disconnect permanently due
to protocol changes. Since the system is quite straightforward, it
usually doesn’t require extra care or monitoring, even without regular
updates. TCP is often referred to as a byte stream, connection-
oriented, transport layer protocol with reliable delivery.
• Byte Stream Delivery: The network layer below and the
application layer above are interfaced by TCP. Applications
use 8-bit byte streams to transfer data to TCP. The sending
Keyword TCP is then responsible for segmenting or demarcating the
byte stream so that the receiver can receive the data in
Internet traffic
is the flow of digestible chunks. The term “byte stream delivery service”
data within the comes from the fact that there are no “record boundaries.”
entire Internet,
or in certain • Connection-Oriented: Before two communicating TCPs can
network links of exchange data, they must first agree upon the willingness
its constituent to communicate. Analogous to a telephone call, a connection
networks.
must first be made before two parties exchange information.
• Reliability: A number of mechanisms help provide the
reliability TCP guarantees.
– Checksums: Every TCP segment has a checksum that
the recipient uses to identify errors in the data or TCP
header.
– Duplicate data detection: Because packets in packet-
switched networks may be duplicated, TCP records the
bytes it receives in order to remove duplicate copies
of data that have already been received.
– Retransmissions: To ensure data delivery in the event of
data loss or damage, TCP must implement retransmission
schemes. Data reception success is verified by the
sender receiving positive acknowledgements from
the recipient. Retransmission is required due to a
timeout period (see timers below) and a lack of positive
acknowledgements.
• Sequencing: Packets may be delivered out of order in packet-
 NETWORK MODEL AND STANDARDIZATION 71
switched networks. TCP’s responsibility is to correctly sequence the segments
it receives so that the byte stream data can be sent to an application in the
correct order.
• Timers: When sending data, TCP keeps track of a number of static and dynamic
timers. Within a set amount of time, the sending TCP waits for the recipient to
respond with an acknowledgement. The sender may retransmit the segment if
the timer runs out before getting an acknowledgement.

2.3.1. TCP Header Format

A TCP segment is the entirety of the TCP header and TCP contained in a single packet.
All valid TCP segments have the format shown in Figure 2.11. The header is 20 bytes
in size when no options are selected. We will quickly go over each field in the provided
TCP header.

Figure 2.11. TCP Header Format (Source: Github, (https://i0.wp.com/www.electronicshub.org/

wp-content/uploads/2013/12/Demultiplexer-Pin-Diagram.png).

2.3.1.1. Source Port

A 16-bit number that indicates which sending host application the TCP segment came
from. There are three ranges for the port numbers: registered ports (1024 through
49151), private ports (49152 through 65535), and well-known ports (0 through 1023).
TCP uses port assignments as an application layer interface. For instance, on TCP hosts,
the TELNET server is always by default assigned to the well-known port 23. A single,
globally unique TCP connection is defined by a complete pair of IP addresses (source
and destination) and a complete pair of TCP ports (source and destination).

2.3.1.2. Destination Port

A 16-bit number identifying the application the TCP segment is destined for on a
receiving host. Destination ports use the same port number assignments as those set
aside for source ports.
72 COMPUTER NETWORKS

2.3.1.3. Sequence Number this segment to the receiving

application as soon as possible.
A 32-bit number identifying the current An example of its use may be to
position of the first data byte in the segment send a Control-BREAK request to
within the entire byte stream for the TCP an application, which can jump
connection. After reaching 2^32 – 1, this ahead of queued data.
number will wrap around to 0.
• Reset the Connection (RST): If this
bit is present, it signals the receiver
2.3.1.4. Acknowledgment Number that the sender is aborting the
connection and all queued data and
A 32-bit number identifying the next data
allocated buffers for the connection
byte the sender expects from the receiver.
can be freely relinquished.
Therefore, the number will be one greater
than the most recently received data byte. • Synchronize (SYN): When present,
This field is only used when the ACK control this bit field signifies that the sender
bit is turned on. is attempting to “synchronize”
sequence numbers. This bit is
used during the initial stages of
2.3.1.5. Header Length connection establishment between
A 4-bit field that specifies the total TCP a sender and receiver.
header length in 32-bit words (or in multiples • No More Data from Sender (FIN): If
of 4 bytes if you prefer). Without options, a set, this bit field tells the receiver
TCP header is always 20 bytes in length. that the sender has reached the end
The largest a TCP header may be is 60 of its byte stream for the current
bytes. This field is required because the size TCP connection.
of the options field(s) cannot be determined
in advance. Note that this field is called
“data offset” in the official TCP standard,
Did you know?
but header length is more commonly used. In May 1974, Vint Cerf and Bob Kahn described an
internetworking protocol for sharing resources using
packet switching among network nodes.
2.3.1.6. Reserved
A 6-bit field currently unused and reserved
for future use. • Window: Data transmission
window size, a 16-bit integer used
by TCP for flow control. The data
2.3.1.7. Control Bits acceptance limit of the recipient
is indicated by this number to
• Urgent Pointer (URG): If this bit field
the sender. Although there is a
is set, the receiving TCP should
“window scale” option that allows
interpret the urgent pointer field.
for even larger windows to be used,
• Acknowledgment (ACK): If this bit the maximum value for this field
field is set, the acknowledgment would limit the window size to
field described earlier is valid. 65,535 bytes.
• Push Function (PSH): If this bit field • Checksum: A TCP sender uses the
is set, the receiver should deliver data fields and TCP header contents
 NETWORK MODEL AND STANDARDIZATION 73
to calculate a value. The same computation will be used to compare this 16-
bit value with the value the receiver produces. The receiver can be extremely
certain that the segment arrived undamaged if the values match.
• Urgent Pointer: Under some conditions, a TCP sender might need to alert the
recipient to urgent data that needs to be handled by the receiving application
right away. When the last byte of urgent data in the segment ends is indicated
to the receiver by this 16-bit field.
• Options: A TCP sender and receiver can use a number of optional parameters
to provide additional functionality. The length of this field will vary depending
on the option(s) selected, but because of the size of the header length field
(4 bits), it cannot be longer than 40 bytes. The maximum segment size (MSS)
option is the most often used option. Using this option, a TCP receiver informs a
TCP sender of the largest segment size it will accept. For different flow control
and congestion control strategies, other options are frequently used.
• Padding: Because options may vary in size, it may be necessary to “pad” the
TCP header with zeros so that the segment ends on a 32-bit word boundary
as defined by the standard.
• Data: Although not used in some circumstances (e.g., acknowledgment segments
with no data in the reverse direction), this variable-length field carries the
application data from TCP sender to receiver. This field, coupled with the TCP
header fields, constitutes a TCP segment.

2.3.2. Connection Establishment and Termination

Over packet-switched networks, TCP offers a connection-oriented service. Connection-
oriented designates that two endpoints are connected virtually. Any virtual connection
goes through three stages. These phases include data transfer, connection termination,
and connection establishment.

2.3.2.1. Three-Way Handshake

In order for two hosts to communicate using TCP, they must first establish a connection
by exchanging messages in what is known as the three-way handshake. The diagram
below depicts the process of the three-way handshake.

Figure 2.12. TCP Connection Establishment (Source: Birkbeck, University of London, https://
www.dcs.bbk.ac.uk/~ptw/teaching/IWT/transport-layer/retransmission.gif).
74 COMPUTER NETWORKS

Three TCP segments are exchanged between two hosts, Host

1 and Host 2, as shown in Figure 2.12. Events in time are depicted
as you read down the diagram. First, Host 1 sends a TCP segment
with the SYN control bit set and an initial sequence number (ISN),
which we represent as the variable x in the sequence number field,
to establish the connection. Host 2 receives this SYN segment at a
later time, processes it, and replies with a TCP segment of its own.
Keyword The SYN control bit is set and its own ISN, represented as variable
y, is contained in Host 2’s response. Additionally, Host 2 sets the
Initial sequence
numbers (ISN) ACK control bit to signal that data beginning with sequence number
refers to the x+1 should be contained in the next anticipated byte from Host 1.
unique 32-bit After receiving Host 2’s ISN and ACK, Host 1 sends Host 2 a final
sequence number
assigned to each acknowledgment segment to complete the connection establishment
new connection process. In this instance, Host 1 sets the ACK control bit and uses
on a Transmission the acknowledgment number y+1 in the acknowledgment field
Control Protocol
(TCP)-based data to indicate the next expected byte from Host 2. Each sender’s
communication. segment includes an exchange of source and destination ports to
use for this connection in addition to the information displayed in
the above diagram.

2.3.2.2. Data Transfer

As long as there is data to send and the sender stays within the
window that the recipient has advertised, a basic TCP implementation
will add segments to the network for a recipient. The receiver
transmits positive acknowledgments back, indicating its location in
the byte stream, as it accepts and processes TCP segments. The
“window,” which indicates how many bytes the recipient is currently
willing to accept, is also included in these acknowledgments. The
byte stream may have a “hole” if data is lost or duplicated.
A receiver that has accepted a byte stream will keep
acknowledging the most recent contiguous place in the stream.
The sending TCP will do nothing but wait for the application to
insert data into the byte stream or for data to arrive from the other
end of the connection if there is nothing to send. The sender must
stop transmission and wait for additional acknowledgments and an
advertised window size greater than zero before continuing if the
amount of data queued by the sender reaches a point where the
amount of data sent will exceed the receiver’s advertised window
size. Timers are employed to prevent unresponsive connections and
deadlocks. Transmissions are delayed in order to increase efficiency.

2.3.2.3. Connection Termination

A connection must be fully closed in four segments in order for it
 NETWORK MODEL AND STANDARDIZATION 75
to be released. TCP requires four segments or host overload. It is important to note
because it is a full-duplex protocol, which that congestion control and flow control
means that each end needs to shut down are distinct concepts. Congestion control
on its own. Figure 2.13 depicts the phase primarily deals with managing network
of connection termination. intermediate devices, such as IP routers,
to avoid overload. TCP’s primary approach
In this scenario, the application to controlling flow is through the window
executing on Host A instructs TCP to close field, as mentioned earlier. The window field
the connection. As a result, Host A and adjusts the flow rate of the byte stream
Host B generate the first FIN segment. between communicating TCPs during data
Upon receiving the initial FIN segment, transfer.
Host B promptly notifies its destination
application of the termination request and In this basic example, a 4-byte sliding
acknowledges the segment. In the event window is utilized. The window “slides”
that Host B’s application also chooses to from left to right as bytes in the stream
terminate the connection, it sends its own are sent and acknowledged. Determining
FIN segment, which Host A processes and the window size and the rate at which to
acknowledges. adjust it is a critical aspect.

2.3.4. Congestion Control

In packet-switched networks, congestion
is a significant problem that can occur.
In communication networks, congestion
is the state in which performance suffers
due to an excessive number of packets in
one area of the subnet. A network may
experience congestion if the load on it
(i.e., packets transmitted to the network)
exceeds the network’s capacity (i.e., the
maximum quantity of packets a network can
manage). Flow control is not the same as
congestion control. One sender outrunning
Figure 2.13. TCP Connection Termination a single receiver (e.g., flow control) is a
(Source: Geeksforgeeks, https://open4tech.com/
data link problem (i.e., a link between
wp-content/uploads/2019/09/session-layer.png).
points). Being a network layer problem,
congestion control addresses what occurs
2.3.3. Sliding Window and Flow when there is more data in the network
than can be sent with acceptable packet
Control delays, no dropped packets, etc. Whereas
The primary objective of the technique congestion control is global, flow control is
known as flow control is to properly align local. Confusion can be avoided by reducing
the sender’s transmission rate with that of the amount of traffic that a specific host
the receiver and the network. A sufficiently sends over the network via a flow control
high transmission rate is essential to ensure mechanism. Congestion is influenced by
optimal performance and prevent network numerous design parameters at the data
76 COMPUTER NETWORKS

link, network, and transport layers. Eliminating the issue is aided

by setting each of these with the goal of avoiding congestion. For
instance, a flooding routing algorithm produces a large number of
redundant packets despite being quick. Overly strict data link layer
timers can result in unnecessary packet retransmissions.

2.3.4.1. Traffic Shaping

Remember
A version of the
The inherent burstiness of computer network traffic is one of the
leaky bucket, reasons for congestion. To smooth out or modify the offered traffic
the generic cell as a function of time is known as traffic shaping. Bursty traffic is
rate algorithm, more difficult to handle than smooth traffic. While a subnet might
is recommended
be able to process 10 million packets in an hour on average, it most
for Asynchronous
Transfer Mode likely won’t be able to process 10 million packets in a minute and
(ATM) networks nothing for the next fifty-nine minutes. The concept of agreeing to
in UPC and NPC a flow specification with a network service provider is related to
at user–network traffic shaping. A flow specification outlines the kind of traffic you
interfaces or
inter-network
plan to send over the internet and the level of service quality you
interfaces anticipate receiving back.
or network-
to-network Some of the quality of service parameters are:
interfaces to
protect a network • loss sensitivity (number of lost bytes per unit of time)
from excessive • loss interval (the unit of time for calculating loss)
traffic levels
on connections • burst loss sensitivity (how long of a loss burst can be
routed through it. tolerated)
• minimum delay (how large of a delay before the app notices)
• maximum delay variation (the variance or jitter in the inter-
packet delay)
• quality of guarantee (how serious the spec is to the
application)

2.3.4.2. Leaky Bucket Algorithm

A buffer that changes an uncontrolled, erratic traffic flow into a
controlled, steady, and predictable flow. A traffic source and the
subnet are placed between the buffer and the source. The buffer
functions as if it were a single server queue with a set maximum
length. Once the buffer is full, packets added to it are dropped.
Either a certain number of packets per unit of time or a total number
of bytes per unit of time (useful if packet sizes vary significantly)
may be drained from the buffer onto the subnet. Token buckets are
a variation of leaky buckets. At a predetermined rate, tokens are
added to the bucket. To exit the bucket, a packet needs to take
hold of and destroy a token. Senders only need to wait for a token
 NETWORK MODEL AND STANDARDIZATION 77
to become available; packets are never lost. Certain applications benefit from bursts,
which are permitted (up to the number of tokens in the bucket) and compromise with
source requirements and congestion.

2.3.4.3. Choke Packets

Using output lines whose utilization has surpassed a certain warning level, routers
send unique choke packets back to hosts while keeping an eye on how their output
lines are being used. By sampling it on a regular basis and combining the new value
(either in use, f = 1, or not in use, f = 0) with the current value for U, the utilization
can be estimated. The weight a determines how quickly the router reacts to changes:
Unew = a * Uold + (1-a) * f
When the sending host receives the choke packet, it reduces the flow of traffic
it is sending to the subnet (often done by factors of 1/2, 1/4, 1/8, etc., as needed).
Then, it stops watching for choke packets, since other packets that are in transit will
probably also cause the congested router to send more choke packets. When a packet
is forwarded on a congested line, it is marked so that downstream routers don’t also
generate choke packets for the sending host corresponding to the same packet.

2.3.4.4. Load Shedding

When routers run out of buffers, they behave as follows. They can try to select the best
packets to discard if needed. This depends on the data link layer error strategy and
the application. Older packets may create gaps in the received packets and could lead
the data link layer to use a go-back-n strategy, which is worse for file transfers than
discarding newer packets. It’s probably better to discard old data (already obsolete?)
and keep new packets for real-time voice or video. Lowering transmission rates for
lower priority packets is necessary to encourage the application to mark packets with
priority. Instead of waiting for congestion to take over, it seems better for routers to
start dropping packets as soon as congestion seems likely.

Did you get it?

1. Which types of control bits used in TCP networking?
2. How to use three-way handshake procedure for connection establishment and
termination?
2.4. USER DATAGRAM
PROTOCOL (UDP)
→Learning Objectives
• Understand the structure and purpose of Protocol Headers.
• Explore the challenges and considerations of implementing TCP over Wireless
Links.
• Identify the key characteristics of Wireless Media.
• Learn about various TCP Performance Enhancements.
• Examine the evolution of Wireless Systems and their impact on TCP.
• Gain knowledge about IP Networks and their functionalities.
• Study techniques for Network Security and Performance Analysis.

One of the main network protocols used to send data over the Internet is the User
Datagram Protocol. It is a component of the Internet Protocol Suite, which sets the rules
for how computers connect to servers and to one another via IP networks. An application
can send datagrams, or messages, to other hosts via the User Datagram Protocol without
the need to first set up specific transmission channels or data paths. Compared to
other methods, this one is less dependable but allows for faster communication. Data
transmission is the main objective of the User Datagram Protocol. It offers no protections
against data corruption during transmission or to guarantee that the information actually
arrives. If error checking is necessary, the protocol assumes that the computers involved
in the communication will take care of it.
A stateless protocol is the User Datagram Protocol. In other words, it handles
every information request as a separate transaction. The benefit of running this way
is that the server can handle more traffic because ongoing conversations don’t require
storage space to be assigned. The User Datagram Protocol is actually superior to more
dependable communication techniques in some circumstances. For example, it enables
the delivery of messages without the lag time brought on by error checking. This is why
time-sensitive applications frequently use it. Voice over IP, a popular streaming media
application, is an example of this kind of application. For servers that process a large
number of tiny requests, the User Datagram Protocol is also helpful. The server can reply
to each request more quickly because it doesn’t have to allocate storage or establish
connections before speaking. All users will experience faster service as a result. For
this reason, servers that host online games frequently use the User Datagram Protocol.
Applications and upper-layer protocols can use UDP as a minimal, erratic, best-effort
message-passing transport. UDP and its variant UDP-Lite are distinct from other transport

78 COMPUTER NETWORKS
 NETWORK MODEL AND STANDARDIZATION 79
protocols in that they do not create end-to-end connections between
communicating end systems. As a result, UDP communication has
little associated end system state and incurs no overhead during
connection establishment or teardown. These features make UDP a
very effective communication transport for some applications, but
they also make it unreliable and lack inherent congestion control.
Applications must be thoughtfully designed because UDP has a
second unique feature: it offers no inherent congestion control. On
many platforms, applications have the ability to send UDP datagrams
at the line rate of the link interface, which is frequently much higher Keyword
than the available path capacity [RFC 4505]. Service access
point (SAP)
UDP is being used more and more frequently as a tunneling is the point at
which a service
protocol. In this scenario, one tunnel endpoint encapsulates the begins (ingress)
packets of another protocol inside UDP datagrams and sends them or ends (egress)
to another tunnel endpoint, which forwards the original packets that and represents
the access point
were part of the payload after decapsulating the UDP datagrams. associated with a
Tunnels can be used to create virtual (private) networks by creating service.
virtual links that, in the physical Internet topology, appear to connect
distant locations directly. Since many middleboxes along the path
support UDP transmissions, using UDP as a tunneling protocol is
appealing when the payload protocol is not supported by them. There
is no communication security offered by UDP. Applications must
thus independently provide security services using extra protocol
mechanisms to safeguard their communications against message
forgery, tampering, or eavesdropping.

2.4.1. Protocol Header

Sending UDP packets doesn’t require a connection to be established
between the computer and the recipient. Due to the fact that a UDP
datagram can only contain a maximum payload of 65,507 bytes for
IPv4 and 65,527 bytes for IPv6, it can only be carried in a single
IP packet. Large IP packet transmission typically necessitates IP
fragmentation. Fragmentation should be avoided since it reduces the
effectiveness and dependability of communication. A computer must
fill in the relevant fields in the UDP header (PCI) before forwarding
the data and header for IP network layer transmission in order to
send a UDP datagram.
The UDP header consists of four fields each of 2 bytes in length:
• Source Port (UDP packets from a client use this as a service
access point (SAP) to indicate the session on the local client
that originated the packet. UDP packets from a server carry
the server SAP in this field)
• Destination Port (UDP packets from a client use this as
80 COMPUTER NETWORKS

a service access point (SAP) to Wireless wide area networks (WWANs)

indicate the service required from like CDPD and GPRS networks, as well as
the remote server. UDP packets wireless local area networks (WLANs) like
from a server carry the client SAP 802.11 networks, can both be represented
in this field) by this model. On the other hand, nodes
• UDP Length (The number of bytes in ad hoc networks can arrange themselves
comprising the combined UDP to create communication routes, and no
header information and payload infrastructure—such as base stations
data) (BSs)—is necessary. Multiple wireless hop
configurations are possible with satellite
• UDP Checksum (A checksum to
networks. Though we also categorize
verify that the end-to-end data has
approaches where this isn’t always the case,
not been corrupted by routers or
we primarily look at configurations where
bridges in the network or by the
the wireless link is the last hop.
processing in an end system. The
algorithm to compute the checksum
is the BN . This allows the receiver 2.4.3. Characteristics of
to verify that it was the intended
destination of the packet because
Wireless Media
it covers the IP addresses, port A number of inherent characteristics of
numbers, and protocol number, wireless media affect TCP performance
and it verifies that the packet is including:
not truncated or padded because
• Channel Losses: While frames are
it covers the size field. Therefore,
being transmitted, signals carried
this protects an application against
by wireless media are susceptible to
receiving corrupted payload data in
substantial interference from other
place of, or in addition to, the data
signals, which results in losses
that was sent. In cases where this
owing to bit modification. Even
check is not required, the value of
with error correction techniques,
0x0000 is placed in this field, in
these losses are challenging to
which case the data is not checked
recover from at the link layer and
by the receiver.
usually necessitate retransmission.
Either the transport layer (TCP)
2.4.2. TCP over Wireless Links or the link layer can handle
TCP makes the assumption that congestion retransmission. Frequent losses at
is the main cause of packet losses. TCP the link layer have an impact on
reduces network utilization in response TCP performance because TCP by
to congestion by shrinking its congestion design assumes that all losses are
window. However, in wireless networks, caused by congestion and calls the
fading and mobility, as well as the high congestion control algorithms when
bit-error rate of the transmission medium, it detects a loss.
can cause losses. TCP continues to respond • Low Bandwidth: Wireless links
to losses in accordance with its congestion may have low bandwidth, which
control scheme, which reduces network occasionally causes the base
utilization needlessly. station to experience excessive
buffering. Due to the possibility
 NETWORK MODEL AND STANDARDIZATION 81
of packets being dropped at the retransmission timeouts to high
base station or being transmitted values, which negatively impacts
back-to-back over the wireless link, TCP performance. Such situations
long observed round trip times may arise in satellite networks and
arise. wireless wide area networks. High
• Signal Fading: When a wireless variance in the measurements is
host is mobile, fading usually takes another factor.
place. Physical interference from
things like weather, obstructions, 2.4.4. TCP Performance
unavailable channels, and areas Enhancements
where multiple cells overlap can
cause signal fading and blackouts.
2.4.4.1. Transport Layer Solutions
Blackouts of this kind may last for
extended periods of time. The primary cause of TCP’s decreased
• Mobility across Cells: Handling performance over wireless links is the
connection handoff is a necessary misinterpretation of wireless losses for
part of a wireless host’s mobility. congestion. As a result, there are many
The base station may retain suggestions for suitable TCP modifications.
connection state regarding the In cellular communications (CC) systems,
transport layer, which may also packets can be lost or delayed during
require handoff, in addition to the handoffs. After the handoff is complete,
link layer state that needs to be recovery from these losses should begin
transferred. Other issues, such as immediately without waiting for a timeout.
the previously mentioned signal TCP is able to accomplish this by getting the
fading, are highly likely to arise right signals from lower layers. Alternatively,
when a host travels between cells. TCP can use lower layer mobility hints to
• Channel Asymmetry: Channel heuristically identify handoff-related losses.
contention resolution is typically TCP can avoid skipping the congestion
asymmetric. More transmission avoidance phase by not halving the slow
time is granted to the sending start threshold during recovery in order
entity than to the receiving entity. to account for these losses. Another
As a result, when channel access method involves the transparent closure
is allowed, TCP acknowledgments of the receiver’s advertised window by
may be queued for transmission wireless link endpoints, which chokes TCP
at the receiving entity’s link layer senders during handoffs. Subsequently, the
and sent back-to-back. This may sender stops all running timers and begins
result in bursty traffic and longer periodically checking the recipient’s window.
round trip times as recorded by the Congestion avoidance aids in assessing
TCP sender, which lower the TCP the new link’s capacity following handoffs.
connection’s throughput. But for other wireless losses, recovery can
• Link Latency: High latencies can be achieved with only retransmissions.
occur on wireless links. When TCP connections can be divided by using
these delays account for a sizable routers connected to both wireless and
portion of the round trip times wired links as pivot points because end-to-
that TCP observes, TCP sets its end retransmissions are slow. This results
82 COMPUTER NETWORKS

in the splitting of end-to-end connections wireless host is transmitting data to a

into distinct TCP sessions for the wired remote receiver; they might even indicate
and wireless portions of the path. Over the congestion losses. In this case, TCP requires
wireless links, a different protocol designed Explicit Loss Notification (ELN) in order to
for error recovery could be used in its differentiate between wireless losses and
place. Since acknowledgments may arrive congestion. The Snoop agent sets an ELN
at the sender before data packets reach bit in TCP headers and propagates it to
their destination, split schemes defy end- the recipient, who then echoes it back to
to-end TCP semantics. Throughput must the sender, if it detects a loss unrelated
be decreased because acknowledgments to congestion. Queue length data can be
must be delayed in order to maintain TCP used by Snoop to heuristically differentiate
semantics. Due to the fact that packets wireless errors from congestion. The TCP
are processed twice by TCP and a sizable sender retransmits the lost packet without
amount of per-connection state is retained, using congestion control after receiving
pivot points incur significant overhead. an ELN notification. ELN works with most
topologies, but it necessitates modifications
to router algorithms. Furthermore, only
2.4.4.2. Link Layer Solutions
after a round trip time has passed and
This is accomplished in CC systems by non- an acknowledgment with the ELN bit set
transparent mode RLPs. Another approach has been returned can a lost packet be
is to carry out IP-level local error recovery, retransmitted.
which is a link layer task, similar to Snoop
While CC system RLPs can retransmit
TCP. By preserving the state of each TCP
data in parallel with TCP, they are not
connection as it passes through a pivot
subject to the layering violations of Snoop,
point, Snoop keeps track of TCP data and
which analyzes TCP headers at the IP level.
acknowledgments. Snoop employs local
However, this is uncommon with completely
timers in conjunction with the loss indications
dependable RLPs, and it is avoided by RLPs
provided by duplicate acknowledgments
that give up on error recovery after a few
to transparently retransmit lost data. It
fruitless tries. Unlike TCP modifications,
does this by caching unacknowledged TCP
link layer schemes operate at the local level
packets. It avoids redundant TCP recovery
with low round trip delays that enable quick
by hiding duplicate acknowledgments from
recovery. Their primary drawback is that
the TCP sender that indicate wireless losses.
they only provide one recovery level, which
Snoop avoids link layer control overhead by
might not be suitable for all higher layer
taking advantage of the data contained in
protocols and applications.
TCP packets. Without going against TCP
semantics, it performs better than split
TCP schemes. Additionally, by suppressing 2.4.5. Wireless System
duplicate TCP acknowledgments, it prevents
incompatible local and TCP retransmissions.
Evolution and TCP
It is necessary for Snoop to locate the TCP Higher speeds and better support for packet
receiver immediately following the pivot data services are the trends for CC systems.
point. The microcells, or small areas with higher
user densities, will have the fastest data
TCP acknowledgments are returned rates available. By allocating multiple TDMA
too late for effective recovery when a slots for every data circuit, the High Speed
 NETWORK MODEL AND STANDARDIZATION 83
Circuit Switched Data (HSCSD) system is a GSM extension that
offers bit rates of up to 56 Kbps. An extension of packet switched
GSM is the General Packet Radio Service (GPRS). Bit rates of up
to 171 Kbps can be achieved with GPRS through dynamic TDMA Keyword
slot reservation. Presently available implementations offer user Explicit Loss
throughput of 20–40 Kbps. There will be about 2% of Internet Notification
(ELN) is a
packet loss, according to experiments. Wideband CDMA is the mechanism by
foundation of UMTS, the third generation European CC system which the reason
that supports both packet switched and circuit switched modes for the loss of a
packet can be
at different bit rates. communicated to
the TCP sender.
Phase one offers GPRS-like services with bit rates as high as
384 Kbps; later phases will offer up to 2 Mbps in certain locations.
The GSM EDGE/IS-136 HS system in the USA will offer bit rates
between 270 and 722 Kbps, or even more than 2 Mbps in certain
places. Many FH spread spectrum systems, like Bluetooth, which
offers bit rates of 400–700 Kbps, are examples of short-range (in-
room) systems, or Personal Area Networks (PANs), that are intended
for low bit rates. Bluetooth’s radio link-level interoperability with
IEEE 802.11 is seriously flawed, even though it should offer TCP
performance comparable to low-end WLANs.
This problem is being worked on by the IEEE 802.15 project,
which defines a PAN standard based on Bluetooth. The Local
Multipoint Distribution System (LMDS) will use the 28 or 40 GHz
frequency bands to provide broadband fixed wireless Internet
access at very high speeds. A Wireless Local Loop (WLL) system
called LMDS gives fixed hosts access to bandwidth between one
and two GHz. We have observed that LMDS can reliably carry TCP
traffic and that it makes use of strong link layer FEC schemes.

2.4.6. IP Networks
Developed in the 1970s, the Internet Protocol (IP) is the main
network protocol used on the Internet. IP and Transmission Control
Protocol (TCP) are frequently used together on the Internet and
numerous other networks; this combination is known as TCP/
IP. IP enables computers connected to a network to have unique
addresses. The Internet Protocol version 4 (IPv4) standard, which
has IP addresses that are four bytes (32 bits) long, is used by the
majority of networks. A 16-byte (128-bit) address is one of the
standard features of the more recent Internet Protocol version 6
(IPv6). Packets are the units of data organization on an Internet
Protocol network. Every IP packet consists of the message data
itself as well as a header that contains information about the source,
destination, and other details about the data. On layer 3 of the
84 COMPUTER NETWORKS

OSI model, IP operates. As a result, it can IP address. Typically, this IP address falls
operate over a variety of data link interfaces, neither at the beginning nor at the end of
such as Ethernet and WiFi. the range of network IP addresses, as the
first and last IP addresses are reserved
Communication between computing for specific purposes. Apart from the
devices is facilitated by networks. Every first and last IP addresses, all addresses
computer (host) connected to a network within the IP network range are considered
must employ the same communication host addresses. Hosts on a network can
protocols in order to communicate with communicate directly with each other
one another. on a one-to-one basis thanks to their IP
A computer network that uses Internet addresses. This type of communication is
Protocol as its communication protocol is known as unicast communication. A host IP
known as an Internet Protocol network. address consists of two parts: a network
Every computer connected to an IP network part and a host part. The network portion
needs to have a unique IP address that of the address identifies the IP network
identifies it as a host. A collection of hosts to which the host belongs, while the host
that utilize Internet Protocol for network part uniquely identifies an individual host.
layer communication and share a physical
connection is referred to as an Internet
Protocol-based network, or IP network. In an Did you know?
IP network, the IP addresses are contiguous, The dominant internetworking protocol in the
meaning there are never any gaps between Internet Layer in use is IPv4; the number 4 identifies
them. the protocol version, carried in every IP datagram.
IPv4 is described in RFC 791 (1981).

2.4.6.1. Address Types in an IP Network

Within a given range of IP addresses used 2.4.6.3. Network Address
in every IP network are special addresses
reserved for: The IP address that appears first in the range
is the network address. More specifically,
• Host Addresses the host portion of the IP address with
• Network Addresses all binary bits set to zero is the address
that is the network address. The network
• Broadcast Addresses
address serves as a means of communication
An IP network also has a subnet mask. between hosts that offer specific network
Each computer has a value stored on it services. In actuality, communication almost
called the subnet mask, which enables it never uses the network address.
to distinguish between IP addresses on an
external network and IP addresses that are
part of the network to which it is connected. 2.4.6.4. Broadcast Address
The final IP address in the range is the
2.4.6.2. Host Address broadcast IP address. More specifically, the
IP address in which every binary bit in
The address of a specific host on an IP the host section of the address is set to
network is referred to as its IP address. one is the broadcast address. A reserved
Each host on a network must have a unique broadcast address enables one host to
 NETWORK MODEL AND STANDARDIZATION 85
broadcast a message to every other host employees from altering any data on the
on the network. This is known as broadcast servers within the company.
communication, and since the last address
in a network is where the host portion is In addition, a systems administrator will
all ones, it is used for broadcasting to all construct a secure firewall for the network
hosts. This unique address is occasionally that might have sentinel software that
referred to as the all hosts address. It is automatically blocks unauthorized programs
possible to designate a different address as from accessing it and an encryption layer.
the broadcast address than the last address Restrictions on websites that might contain
with certain vendors. Keep in mind that malicious coding or malware that installs
every network has unique broadcast and itself on a user’s computer can also be
network addresses. The host component of applied by the administrator to employees’
a network address does not always have a computers. Programs to remove malware
zero in its decimal representation. and adware are available for both networks
and individual computers. Flexibility is one
issue that typically comes up with this
2.4.7. Network Security and kind of security. Management must strike
Performance Analysis a compromise between security concerns
and workers’ need to access websites for
The goal of network security is to prevent work.
unwanted access to the underlying
computer network infrastructure. The It takes communication between
network administrator is in charge of it, and management, the systems administrator,
it involves ongoing effectiveness assessment and staff for network security to function
and monitoring. and for staff members to be able to use it.
Security lapses are expensive and harmful
Every business that uses computers to output and efficiency, as demonstrated
should be very concerned about network by the experience with prior network and
security, which is the set of precautions a email viruses. U.S. Businesses invest millions
company takes to safeguard its computer of dollars annually in network security
system. A hacker or rival with access to solutions. A multi-pronged attack is the
sensitive or important data could cause data best defense a company can have against
loss or even the system’s total destruction breaches. A network can be kept secure
if network security is compromised. A with the help of firewalls that prevent
network has reached appropriate security single points of access, astute systems
when a user must pass through multiple administrators, regular security updates,
security levels in order to gain access and early anti-adware software installation.
to the desired network. A system is
more secure the more layers it contains. The different wireless network security
Because he has administrator rights on methods are:
the system, the systems administrator is
frequently in charge of network security. 2.4.7.1. Wired Equivalent Privacy (WEP)
Actually, the mainframe server and any
associated computer terminals should This technique encrypts data as it travels
only be accessible administratively by the between computers on the network using a
systems administrator and his helpers. This key that is set. This key is needed during
will assist in preventing any unauthorized wireless network setup in order to encrypt
86 COMPUTER NETWORKS

and decrypt sent data. This is a comparatively outdated method of network security,
and it is not too difficult to crack the WEP key.

2.4.7.2. 802.1X Authentication

Particularly, 802.1X is utilized for both wired and wireless 802.11 networks. Most often,
networks in offices use this kind of authentication. When a user requests access to the
network, their identity is verified by an authentication server.

2.4.7.3. Wi-Fi Protected Access (WPA)

The most popular approach to secure wireless networks is this one. Information is
encrypted and only those with permission can access the network when using Wi-Fi
Protected Access. This encryption method has been updated and is known as WPA2.
Because it is more secure than WPA, it is advised to use it. Nevertheless, older wireless
devices (such as MP3 players) and access points might not support WPA2.

Did you get it?

1. What are the various TCP Performance Enhancements?
2. How does the evolution of Wireless Systems impact TCP?
 NETWORK MODEL AND STANDARDIZATION 87

A CLOSER LOOK
The network model serves as a foundational framework for communication and data
exchange within computer systems. It delineates how devices within a network interact,
communicate, and share resources. At its core, the network model comprises various
layers, each responsible for specific tasks such as data transmission, addressing, and
error detection. This modular approach ensures efficiency, scalability, and interoperability
across heterogeneous network environments.
Standardization plays a pivotal role in the development and implementation of
network models. By establishing uniform protocols, interfaces, and technologies,
standardization facilitates seamless communication between diverse devices and systems.
Key organizations such as the Institute of Electrical and Electronics Engineers (IEEE) and
the Internet Engineering Task Force (IETF) play instrumental roles in defining and refining
networking standards. These standards encompass a wide array of aspects, including
data formats, communication protocols, security measures, and network topologies.
The network model and standardization are intricately intertwined, with standards
providing the guidelines and specifications for implementing various layers and protocols
within the network model. For instance, the OSI (Open Systems Interconnection) model,
a prominent network model, delineates seven layers, each governed by standardized
protocols and interfaces. These standards ensure that devices from different manufacturers
can communicate effectively, fostering compatibility and interoperability.
Moreover, standardization fosters innovation and market growth by providing a common
framework upon which developers can build new technologies and services. It enables
seamless integration of diverse networking technologies, fostering the development of
sophisticated applications and services.
88 COMPUTER NETWORKS

ROLE MODEL
Christopher S. Strachey: One of the founders of denotational semantics,
and a pioneer in programming language design and computer time-
sharing.

Biography
Born on November 16, 1916 into one of England’s more prominent
families, Christopher Strachey was educated at Gresham’s School,
Norfolk, and at King’s College, Cambridge. He spent the war years
in radar research, after which he spent several years as a schoolmaster. His career in
computing did not effectively begin until 1951 when he started to program the machines
at the National Physical Laboratory and Manchester University, while still a master at
Harrow School.
In 1962, he also began work in the University Mathematical Laboratory, Cambridge,
where, in the nominal position of “part-time research assistant,” he worked on the
CPL programming language. In 1965, he wound up his consultancy and resigned from
Cambridge in order to form the Programming Research Group at Oxford University.
Under his leadership, the group quickly established an international reputation in the
theory of programming. At Oxford, in collaboration with Dana Scott, he produced the
work of which he was most proud, the foundation of denotational semantics; this work
was just beginning to bear fruit when he died in 1975.
Returning to King’s in October 1938 for a fourth and final year, he graduated in
the summer with a “lower second” in the natural sciences tripos. This mediocre result
was a considerable disappointment, and it dashed any hopes he had entertained of a
research studentship. Obliged to find an occupation of some kind, he accepted a post
as a physicist with Standard Telephones and Cables Limited (STC) at a salary of £4
per week.
Strachey began work in the Valve Development Laboratories of STC, London, in
August 1939, just a month before the declaration of war with Germany.
Most of his time at STC was spent as one member of a small team led by J.H.
Fremlin (later a professor at the University of Birmingham), investigating the theoretical
design of centimetric radar valves. Strachey’s particular contribution was the derivation
of analytical formulas for valve parameters and their experimental verification. His
mathematical work involved the integration of differential equations, some of which
proved particularly intractable; so, with colleague P.J. Wallis, he began to obtain
numerical solutions using a differential analyzer. He later came to regard this experience
with a computing machine as being something of a turning point, and his interests in
computing were aroused generally at this time; he began to read the literature on the
subject, and from time to time he also assisted STC colleagues in computational tasks.
 NETWORK MODEL AND STANDARDIZATION 89

In July 1944 Strachey left Fremlin’s group and was transferred to London to work
in the STC Radio Division. This work was concerned with electrical and mechanical
design, which he found much less to his taste than the theoretical work. Strachey had
never liked the atmosphere of the STC laboratories (either at Ilminster or at London),
which he found “rather narrow-minded and sordid.”
Strachey began as a “physics-cum-mathematics” master at St. Edmund’s in October
1945, at a salary of £335 per annum.
He seems to have been a thorough schoolmaster: his lessons were meticulously
prepared, and he spent much effort coaching his pupils for public examinations. After
many applications to other schools, he was finally offered a position at Harrow School
and left St. Edmund’s in the spring of 1949.
Strachey began teaching at Harrow School in September 1949, at a salary of £600
a year. Securing a post at Harrow, one of Britain’s leading public schools, was a real
advancement to his career.
Strachey spent many of his evenings organizing societies and clubs. He was particularly
keen on the science society and gave talks on topics such as interplanetary travel and
the surface tension of soap films. He also played bassoon in the school orchestra, an
instrument he had taught himself at St. Edmund’s.
Against this background of busy “schoolmastering,” Strachey’s interest in computing
had been largely dormant since leaving STC. Of course, computers were very much in
the air in the late 1940s, and he no doubt read the semipopular articles.
His first exposure to a stored-program computer occurred in January 1951 when,
through a mutual friend, he obtained an introduction to Mike Woodger of the National
Physical Laboratory (NPL).
Back at Harrow, he began to write a program to make the Pilot ACE play draughts.
This was typical of his early attempts at programming: anyone with more experience
or less confidence would have settled for a table of squares. He got the idea of using
the machine to play draughts largely from an article by Donald Davies of NPL, “A
Theory of Chess and Noughts and Crosses,” that had appeared in the June 1950 issue
of Penguin Science News.
The following spring, Strachey learned from Woodger of the Ferranti Mark I computer
that had just been installed at Manchester University. This machine had a much larger
store than the Pilot ACE, with correspondingly greater scope for Strachey’s kind of
programming. Alan Turing, who was then assistant director of the Manchester University
Computing Machine Laboratory, had written the programmer’s handbook for the machine,
and Strachey had known him just well enough at King’s College that he could ask for
and receive a copy.
He visited the Manchester Mark I for the first time in July 1951. When Strachey
explained his ideas for a draughts-playing program, Turing was much impressed and
90 COMPUTER NETWORKS

suggested that another interesting problem would be to make the machine simulate
itself, in the fashion of the interpretive trace routines developed for the Cambridge
University EDSAC. Strachey was attracted by this idea and temporarily put the draughts
program to one side. The final trace program was some 1,000 instructions long-by far
the longest program that had yet been written for the machine, although Strachey was
unaware of this.
Shortly after, Strachey received a letter from M.H.A. Newman, professor of pure
mathematics at Manchester, complimenting him on the quality of his programs (which
had been reported by Turing), and saying that he hoped to be able to offer him a post
in the laboratory when one became available. Before he could do so, however, Strachey
came to the notice of Lord Halsbury, managing director of the National Research and
Development Corporation (NRDC). In November 1951 he was formally offered a post
as technical officer with NRDC at a salary of £1200. Strachey formally began as an
employee of NRDC on June 3, 1952.
During the first week of September 1952, Strachey attended the second ACM
National Conference, held at the University of Toronto. The conference was timed to
coincide with the inauguration of the FERUT, the second Ferranti Mark I, which had
been installed in the Computation Centre of the university.
NRDC had agreed to loan Strachey to the University of Toronto to help with
programming the calculations for the St. Lawrence Seaway project. Strachey spent all of
October and November, and part of the following spring, on the program. The program
was very long-about 2,000 instructions-and the input data tape was estimated to be
one and a half miles in length.
During his time in North America, Strachey visited various research labs and computer
manufacturers in the US, conducting a detailed study of different computers’ order codes.
He was one of the first in Britain to recognize the importance of this subject. In early
1954, he began discussions with Ferranti about designing a new machine, the Ferranti
Packaged-Circuit Computer (FPC, later known as Pegasus). He convinced Ferranti to
establish a small department to create a programming system for the machine, actively
participating in its development. The Pegasus programming system was a significant
achievement in British programming during the 1950s and had a profound impact.
Strachey, though relatively new to computing, was 38 years old when he started
working on Pegasus, displaying great energy and maturity. The project thrived under
his leadership, which was never as strong in subsequent endeavors.
In the late 1950s, multiprogramming (also referred to as time-sharing) gained
popularity. Strachey’s concept of time-sharing aimed to maintain direct interaction between
programmers and machines while benefiting from the efficiency of multiprogramming. He
applied for a time-sharing patent in February 1959, causing some tension with Kilburn’s
group at Manchester due to similarities with the Atlas project. In March 1959, after
eight years at NRDC, Strachey announced his decision to resign and pursue freelance
 NETWORK MODEL AND STANDARDIZATION 91

director and consultant roles. Strachey formally started activities as a private consultant
on June 1, 1959, operating from his private address of 9 Bedford Gardens, Kensington.
As part of his consultancy agreement with Ferranti, Strachey undertook to deliver a
scientific autocode for the new ORION computer. To do this work, he took on a full-
time employee, Peter Landin, in January 1960. Landin spent only part of his time on
the autocode; with Strachey’s encouragement, he spent the remainder of his time on a
theoretical study of programming languages. It gave Strachey a certain ironic satisfaction
that he was financing “the only work of its sort being carried out anywhere (certainly
anywhere in England)” (Strachey 1971). Landin’s work, which concerned the application
of Church’s lambda calculus to programming language semantics, was described in a
classic paper, “The Mechanical Evaluation of Expressions.”
As well as financing Landin’s theoretical studies, Strachey was also prominent in the
public debate on programming languages, particularly Algol. For example, at the 1959
UNESCO conference in Paris, he was in fine form, holding forth on the deficiencies of
Algol while perched on the edge of a table on the platform, his legs swinging.
Strachey became well known for his outspoken view that Britain had fallen seriously
behind the US in the field of programming. The main reason for the poor progress in
Britain, Strachey perceived, was the small size of British machines. It was simply not
possible to implement a LISP compiler on a machine the size of Pegasus, which was
then the workhorse of most university computer centers.
In June 1962, Wilkes invited Strachey to work full-time in the University Mathematical
Laboratory, Cambridge, to participate in the development of a new programming language
and compiler for the Titan computer. Strachey accepted enthusiastically and began work
the following month. A considerable financial sacrifice was involved, for the salary that
Wilkes was able to offer was very modest compared to the consultancy fees he was
then able to command.
The CPL (Cambridge Programming Language) project got off to a most promising
start. By mid-August, the project group, which consisted of Strachey, Baron, and David F.
Hartley, had produced an outline proposal. In the autumn, the group began to collaborate
with Eric Nixon and John N. Buxton of the London University Computer Unit, so that
CPL could also be used on the London University Atlas.
Concurrent with the CPL activity at Cambridge, it must be recalled that Strachey
was also in business as a private consultant, with an office, a secretary, and his
principal assistant Peter Landin to support. From mid-1963, he was actively seeking an
opportunity to lead a university research group in the theory of programming. At first,
it was hoped to set up a research unit in the Mathematical Laboratory at Cambridge,
but this proved unsuccessful. In January 1964, Strachey also unsuccessfully competed
for the chair of Computing Science at Imperial College, University of London. Finally,
he secured the support of Leslie Fox, who succeeded in obtaining a DSIR grant to set
up a programming research group at Oxford University in July 1965.
92 COMPUTER NETWORKS

Although DSIR support for the Programming Research Group (PRG) started in July
1965, Strachey didn’t move to Oxford until April 1966. The work on CPL continued. In
June, Strachey gathered the full CPL group for a meeting, and it was decided, probably
due to his enthusiasm for revising earlier drafts, that Strachey should edit and distribute
the final version of the CPL reference manual. Throughout the rest of 1966, Strachey
and David Park kept working on CPL and preparing the reference manual, titled the
CPL Working Papers.
In the latter half of 1967, Strachey penned “Fundamental Concepts in Programming
Languages,” one of his most significant and lengthy papers, which, as usual, remains
unpublished in its original form. Like much of Strachey’s work at Oxford, though, the
paper had a significant impact through private circulation.
Since starting at Oxford, Strachey had limited his external activities to focus on his
academic work. For instance, he stepped down from his directorship of CAP in 1967,
although he remained a shareholder and a consulting fellow. He became less involved in
British Computer Society matters. He did participate in the computer science subcommittee
of the SRC and gave occasional external lectures and did some consulting, but these
activities were much less frequent compared to the early 1960s. On one memorable
occasion, he collaborated with Stanley Gill and Alex d’Agapeyeff to address a Labour
Party Science and Technology Committee, and in July 1973, he took part in a television
debate on the Lighthill report on AI research in Britain.
Strachey’s research into mathematical semantics took a significant step forward
in 1969 through his collaboration with Dana Scott. Scott, a mathematical logician at
Princeton University at the time, was on a sabbatical leave in Europe when he first
met Strachey at a lively meeting of IFIP Working Group 2.2 in Vienna during Easter.
Scott’s contribution was to provide a sound mathematical basis for the lambda-
calculus models that Strachey used in his formal semantics. Strachey had first used
this device in 1964 in his paper “Towards a Formal Semantics,” but now conceded it
was “gravely lacking in mathematical rigour.”
Strachey decided, early in 1973, to submit an essay for the Adams Prize of Cambridge
University, an award that has secured the reputation of many distinguished British
mathematicians over the years. The subject for the Adams Prize for 1973-1974 was
“Computer Science excluding Hardware,” and the competition was open to persons
admitted to a degree of the university. Apart from its obvious practical value as the kind
of tangible evidence that would impress a Royal Society election committee, winning
the Adams Prize perhaps meant something deeper to Strachey in terms of recognition
from the alma mater. The essay for the Adams Prize dominated the remainder of his life.
The Adams essay included an important historical account of the development of the
Oxford semantics and a final reworking of his “Fundamental Concepts in Programming
Languages.” The essay, with a few sections still unwritten, was finally dispatched to
the Cambridge University Registry at the very end of 1974. The effort of writing the
Adams essay took its toll on Strachey, and early in 1975 he spent several weeks away
 NETWORK MODEL AND STANDARDIZATION 93

from Oxford resting. He returned to Oxford in the spring, where he continued to work
with Milne at completing the essay and revising it into book form. In a matter of weeks,
however, he contracted an illness diagnosed as jaundice; he obeyed the usual dietary
restrictions and made an apparent recovery, but the illness quickly returned. He died
of infectious hepatitis on May 18, 1975. The winner of the Adams Prize was announced
shortly after Strachey’s death. The submission of Strachey and Milne did not win.
Strachey made three important technical contributions to computing in Britain: the
logical design of computers, the design of programming languages, and the development
of denotational semantics. Strachey was responsible for a strong current of influence
in the design of programming languages. Perhaps more important than any details of
a particular language was the example that Strachey set: his whole approach to the
subject and his way of going about things. Donald Michie made this point well when
he wrote in 1971: Today an “invisible college” of programming theory exists throughout
the Universities of Britain. Almost every member of this “college” was guided along the
path at some stage by Strachey’s direct influence. Developments of theory may in the
long run prove decisive in helping to clear the hurdles of software engineering which
still lie ahead (Michie 1971).
94 COMPUTER NETWORKS

SUMMARY
• A network’s foundational component is its networking software. Administrators
can use it to deploy, maintain, and keep an eye on networks. Conventional
networks consist of specialized hardware, like switches and routers, which
combine networking software into the whole setup.
• The ISO/OSI model, which stands for International Standards Organization/
Open System Interconnection, is a commonly used reference model for network
communication between two end users. To solve future issues, it can be useful
to have a fundamental understanding of the network’s operation.
• Transmission of data over the network is handled by the transport layer. The
data is conceptualized at this level more as a conversation than as individual
packets. This is achieved by using protocols, which are referred to as “rules of
communication.”
• The network layer makes sure that every packet makes it from its point of
origin to its destination, while the data link layer manages the packet delivery
between two systems connected by the same network (links).
• The physical layer of a network consists of the actual cables, fibers, cards,
switches, and other electrical and mechanical components. This layer converts
digital data into signals so that they can be transmitted over a wire.
• Multiplexing is the process of combining several signals (analog or digital)
for transmission over a single line or medium. Multiple low-speed signals are
combined for transmission over a single high-speed connection in a common
form of multiplexing.

MULTIPLE CHOICE QUESTIONS

1. Which organization developed the OSI (Open Systems Interconnection) model for
network architecture?
a. IEEE
b. ISO (International Organization for Standardization)
c. IETF (Internet Engineering Task Force)
d. ITU (International Telecommunication Union)
2. Which layer of the OSI model is responsible for providing end-to-end communication
between applications?
a. Application Layer
b. Transport Layer
c. Network Layer
d. Data Link Layer
 NETWORK MODEL AND STANDARDIZATION 95
3. Which TCP/IP model layer corresponds to the OSI model’s Physical and Data Link
layers?
a. Network Access Layer
b. Internet Layer
c. Transport Layer
d. Application Layer
4. Which organization is responsible for the development of TCP/IP standards and
protocols?
a. IEEE
b. IETF (Internet Engineering Task Force)
c. ISO (International Organization for Standardization)
d. ITU (International Telecommunication Union)
5. Which of the following is not a characteristic of the TCP/IP model?
a. It is a hierarchical model.
b. It is based on a set of protocols developed for the ARPANET.
c. It consists of four layers: Application, Transport, Internet, and Network Access.
d. It was developed before the OSI model.

REVIEW QUESTIONS
1. What mechanisms are used in the transport layer?
2. How is framing detection understood in the Data Link Layer?
3. What are the various types of multiplexing?
4. What is the purpose and functionality of digital multiplexers?
5. How is connection establishment and termination clarified?

Answer to Multiple Choice Questions

1. (b) 2. (a) 3. (a) 4. (b) 5. (c)

REFERENCES
1. Alderson, D., Li, L., Willinger, W., & Doyle, J. C., (2005). Understanding internet
topology: Principles, models, and validation. IEEE/ACM Transactions on Networking,
13(6), 1205–1218.
2. Bhandarkar, S., Jain, S., & Reddy, A., (2005). Improving TCP performance in high
bandwidth high RTT links using layered congestion control. Proceedings of the 3rd
International Workshop on Protocols for Fast Long-Distance Networks.
96 COMPUTER NETWORKS

3. Bresnahan, J., Link, M., Khanna, G., Imani, Z., Kettimuthu, R., & Foster, I., (2007).
Globus GridFTP: What’s new in 2007. Proceedings of the 1st International Conference
on Networks for Grid Applications. Article 19.
4. Kelly, T., (2003). Scalable TCP: Improving performance in highspeed wide area
networks. ACM SIGCOMM Computer Communication Review, 33, 83–91.
5. Leith, D., & Shorten, R., (2008). Next generation TCP: Open questions. Proceedings
of the 6th International Workshop on Protocols for Fast Long-Distance Networks.
6. Magoni, D., & Pansiot, J. J., (2002). Internet topology modeler based on map
sampling. Proceedings of the 7th International Symposium on Computers and
Communications (ISCC 2002) (pp. 1021–1027).
7. Mahadevan, P., Hubble, C., Krioukov, D., Huffaker, B., & Vahdat, A., (2007). Orbis:
Rescaling degree correlations to generate annotated internet topologies. ACM
SIGCOMM Computer Communications Review, 37(4), 325–336.
CHAPTER
Local Area
3 Network/Wide
Area Network

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Prepare for the OSI model

• Know about the multiplexing and switching
• Discuss on transmission control protocol (TCP)
• Learn about the user datagram protocol (UDP)
98 COMPUTER NETWORKS

INTRODUCTORY EXAMPLE

LOCAL AREA NETWORK DESIGN AND CONFIGURATION IN

AN ORGANIZATION
When it comes to setting up a Local Area Network (LAN) for an organization, it’s all
about taking a systematic approach to meet the needs of users while ensuring security,
reliability, and scalability. LANs are crucial for internal connectivity, data exchange,
resource access, and seamless communication within the organization. A well-thought-
out LAN design involves considering elements like network topology, hardware selection,
addressing schemes, security measures, and management protocols. Choosing the
right topology is key in LAN design as it dictates how devices are connected within
the network. Common topologies include star, bus, ring, and mesh. In organizational
setups, the star topology is often favored for its centralized management and ease
of troubleshooting. Each device connects to a central switch, enabling efficient data
transmission and network management.
Hardware selection is vital in LAN configuration, with switches, routers, access points,
and cabling infrastructure forming the network’s backbone. High-performance switches
with adequate port density and throughput are essential to meet the organization’s
bandwidth needs. Routers facilitate interconnectivity between LANs and provide access
to external networks, while access points enable wireless connectivity for mobile devices
and remote access.
Addressing and subnetting schemes are essential for efficient resource allocation and
network management. Implementing a structured IP addressing scheme using IPv4 or
IPv6 addresses helps organize devices logically and simplifies troubleshooting. Subnetting
enables the segmentation of the network into smaller, manageable subnets, improving
performance and security by controlling traffic flow and minimizing broadcast domains.
Security measures are paramount in LAN configuration to safeguard sensitive data
and protect against unauthorized access or malicious activities. Implementing firewalls,
intrusion detection/prevention systems, and access control lists (ACLs) helps enforce
security policies and restrict access to authorized users only. Encryption protocols such
as WPA2/3 for wireless networks and VPNs for remote access ensure data confidentiality
and integrity.
Scalability is a crucial consideration in LAN design, as organizations need networks
that can accommodate growth and expansion over time. Designing a scalable LAN
involves deploying modular hardware, implementing redundant links for high availability,
and adopting protocols such as Spanning Tree Protocol (STP) to prevent network loops
and ensure stability.
INTRODUCTION
A computer network that is located within a constrained physical space, such as a
house, school, office building, computer lab, or collection of buildings, is known as a
local area network, or LAN. Workstations and personal computers that are networked
together to form a local area network (LAN) can access and share data and devices,
including printers, scanners, and data storage devices, from any location on the LAN.
Higher data transfer and communication rates, as well as the absence of the need for
leased communication lines, are characteristics of local area networks (LANs). Digital
data is sent over serial transmission lines by local area networks. Because it allows for
relatively high transmission speeds by sending character blocks at a time, synchronous
transmission is typically used.
A wide area network, or WAN, is a communications system that connects many
different locations, like cities, states, or nations. They can be more public to link smaller
networks together, or they can be more private to connect sections of a business.
The simplest way to comprehend what a wide area network (WAN) is to consider the
internet as the largest WAN in the world. Because it connects numerous smaller local
area networks (LANs) or metro area networks (MANs) via the use of ISPs, the internet
is a wide area network (WAN). A company may, on a smaller scale, have a WAN that
connects its headquarters, smaller branch offices, and cloud services. In this scenario,
the WAN would be utilized to link all of those corporate divisions.
The ultimate goal is always to enable communication between various smaller
networks located in different places, regardless of what the WAN connects or how far
apart the networks are from one another. A wide-area network, or WAN, is one that
spans a sizable geographic area. A wide area network (WAN) links several smaller
networks, such as metropolitan area networks (MANs) and local area networks (LANs).
This guarantees that users and computers in one place can communicate with users and
computers in other places. Either a private network or the public transmission system
can be used to implement a WAN.

LOCAL AREA NETWORK/WIDE AREA NETWORK 99

100 COMPUTER NETWORKS

3.1. LOCAL AREA

NETWORK
→Learning Objectives

• Trace the historical development and milestones of local area networks (LANs).
• Identify different types of LAN architectures and topologies.
• Analyze the role of LAN applications in enhancing productivity and efficiency
in business environments.
• Define the characteristics and attributes that distinguish LANs from other types
of networks.
• Identify the different types of devices commonly found in LAN environments.
• Explore the hardware components and infrastructure required to establish and
maintain a LAN.
• Analyze the advantages and benefits of deploying a LAN in a computing
environment.
• Explain the importance of protocol standards in ensuring interoperability and
compatibility in LAN environments.
• Explore the standards and specifications that govern LAN technology and
operations.

The network is privately owned and hardware such as a printer. For computer
is known as the local area network. A communications, a local area network, or
network is an assembly of computers and LAN, is a type of shared packet network
other devices that are linked together to with limited distance. Every machine in a
exchange data. A local area network, or LAN is linked by a single cable. A LAN
LAN, is a type of network that can function can have data rates as high as 100 Mbps,
over a small physical area, such as a few with a range of 4 to 16 Mbps. A local area
kilometers in size, which could be an office, network (LAN) can also simply refer to the
factory, or collection of buildings. Widely hardware and software that enable device
used in a range of computers, LANs allow networking. In this regard, Ethernet and
resources (e.g., printers) and communication Local Talk are two different types of LANs.
of data. To share resources, LANs connect The Local Talk protocol is called AppleTalk.
the workstations and personal computers
in the offices. The resources that need Hardware, protocols, and cabling
to be shared can be data, software, or standards comprise the parts that LANs
use. Ethernet, Token Ring, TCP/IP, 5MB,
 LOCAL AREA NETWORK/WIDE AREA NETWORK 101
NetBIOS, NetBeui, IPX/SPX, Fiber Distributed Data Interchange (FDDI), and Asynchronous
Transfer Mode (ATM) are some examples of LAN protocols.
LANs can be identified from other kinds of networks based on their topology, size,
and transmission technology. A limited local area network (LAN) is one that can cover a
comparatively small area. LANs employ cable-based transmission technology. Conventional
LAN technology has very low error rates and can transmit data at 10 Mbps to 100
Mbps. But LAN technology today can transfer data at up to 10 Gbps. For broadcast
LANs, a variety of network topologies are possible. One client machine is the master
machine in a linear cable network (Bus) at any given time, and it is only permitted to
transmit. Every other machine doesn’t send any data. The ring is the second kind of
broadcast LAN. There, LAN speeds of 4 and 16 Mbps are used. Another illustration of
a ring network is FDDI.

3.1.1. The Evolution of LAN

The office setting makes extensive use of starts to become limited. Gigabit Ethernet
PCs and desktop computers, so it became cards can now be purchased. The 10 Mbps
clear that the cost of connecting a printer cards have been superseded. Currently, the
or FAX machine to every single computer market offers 10/100, 10/100/1000 auto-
would be prohibitive. It also takes time sensing cards. The future of autosensing
to copy files to a disk and transfer them is dual-mode technology. Despite its
between computers in order to print them. increasing popularity, Gigabit Ethernet
Significant savings resulted from connecting LANs are ten times faster than 100Mbps
computers so they could share a printer LANs. ALOHA was the original protocol
and files. When 10 Mbps Ethernet cards from which Ethernet evolved. ALOHA was
first appeared in 1982, they were highly primarily utilized in satellite-based packet
priced. Even in 1988, 10 Mbps Ethernet radio networks for communication. To put
was still functional in small installations it briefly, an earth station sends some data
and offered respectable performance for and then waits for an acknowledgment
large LANs. Congestion started to appear (ACK) when the data is ready. It will time
on large installations by 1990. As a result, out and send the same message again if
alternatives to 10 Mbps cards were gaining it does not receive an ACK. Up until the
traction. Installing a switched architecture transmission is successful, the sender never
was one of these options instead of building gives up. Determine the maximum number
a hub-based architecture. of users to guarantee an area with the aid
of traffic analysis.
For smaller installations, 10Mbps
switched LANs were offering serviceable
performance in 1996. At this point, shared 3.1.2. LAN Technology
Ethernet at 100 Mbps was an incredible
In the information age, local area networks,
technology. By 1998, large LANs and
or LANs, are now widely used. The advent
small LANs alike were using switched
of low-cost computers and the requirement
100 Mbps and switched 10 Mbps Ethernet,
for hardware resources and information
respectively. As more people start using
sharing in an office setting paved the way
video conferencing on large LANs, 100 Mbps
for the development of low-cost local area
102 COMPUTER NETWORKS

networks (LANs). LAN technologies are intended to link computers

together over short distances (e.g., inside one structure or space).
Modern LANs are cheap, incredibly dependable, and simple to set up
and maintain. An interface card is required for a computer in order
to connect to a LAN. This interface card is seen by the computer
Keyword as any other I/O device. When data is transmitted to the interface
Communication card, it is received by another computer connected to the same
medium is a communication medium (copper wire, coax cable, optical fiber, or
channel or route in through the air) at a specific radio frequency. The application can
the communication
process through see through the communication channel that is being used.
which a sender
communicates their Because LANs have altered the economics of computing, they
message to the have become incredibly popular. Cheap computers can now be used
receiver.
to share access to shared resources like printers, large storage
banks (disks), and application software. LANs facilitate group
and collaborative work by allowing users to share information.
Ethernet is arguably the most widely used LAN technology. Ethernet
communicates between various LAN stations via a broadcast
mechanism. When a station wants to talk to another station, it
just listens to the channel and, if no one is talking, sends its
data onto the communication medium along with the station’s
numerical address. Each station scans the communication medium
for information transmitted with its address. If it detects such
information, it receives it and forwards it to the relevant application
program.

3.1.3. LAN Applications

Local Area Networks can be utilized to provide any or all of the
following facilities (see Figure 3.1):
1. The interconnection of all computers, terminals, PCs, and
other workstations within a department or building.
2. The effective communication between all devices attached
to the network, regardless of their vendor.
3. The sharing of expensive peripheral resources attached
to the network such as disks, printers, plotters, central
processors, and databases.
4. The standardization of wiring, hardware, and software
required for communication and applications.
5. The establishment of a foundation for continued growth
and the expansion of a distributed computing environment.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 103
categories according to the kind of
service they provide.
• File server and disk server: A file
server provides different client
pieces with disk storage space
(of several gigabytes) in response
to data requests from application
programs running on client
workstations. When multiple clients
request data at the same time, it
also maintains data consistency.
Any application running on a
workstation within a LAN that
uses a file server can request a
file involving the local operating
system. The server receives the
Figure 3.1. Example of a typical LAN file request from the workstation
application (Source: Quora, https:// and processes it before sending
qph.cf2.quoracdn.net/main-qimg- the necessary file straight to the
9682d6e2cd0628b9e7ddb061c8237166-pjlq). workstation. File servers and
disk servers are comparable. The
distinction lies in the fact that
3.1.4. Characteristics of a LAN when an application operating on
A LAN can be characterized by means of a workstation makes a request for
hardware and software components and a a particular file on a file server,
set of protocols. the workstation’s local operating
system communicates with the
Hardware components of a LAN are:
file server to obtain the necessary
• The Server: A server is a computer data. The application can access
on a network that offers services the necessary file directly from a
to other workstations and clients. disk server without the help of the
Data management is a LAN server’s local operating system.
main objective. It safeguards, keeps, • Database server: A subset of file
and retrieves the data. A server servers make up these servers.
also transmits data to authorized They give their clients access to
remote users and requesters on enormous databases. Data from
the network. The type of server databases is kept on optical, CD-
and its configurations are primarily ROM, and hard disk drives. The
determined by the reason behind DBMS, which is more advanced
the LAN’s construction. Depending than the straightforward file I/O
on the kind of service they offer, access technique, is contained in
servers can be categorized. it. Data redundancy is removed by
Numerous services could also be DBMS, enabling transparent data
offered by a single server. The distribution for the user. Unlike a
servers are listed below in different file server, which sends the entire
104 COMPUTER NETWORKS

file to the requesting client, a of workstations include Microsoft

database server only extracts the Windows NT Workstation, Windows
pertinent data and sends it. 2000 Workstation, IBM Workstation,
• Print server: Each workstation Sun Workstation, and so on.
or node on a LAN shares one or • The Transmission Media for LAN:
more printers. The job of a print To build a LAN, a variety of
server is to collect data from transmission media is used. Optical
multiple workstations, store it on fibers, coaxial cables, and twisted
disk, and send it to the printer. pairs are the most widely used
Print spooling is the term for this types. Wired LANs are those that
process. Nowadays, print services use any type of such transmission
are integrated into file servers. Any media. It is also possible to build
PC connected to the network can LANs without using cables or
act as a print server in many LAN transmission media. These LANs
architectures. are referred to as wireless LANs.
• Gateway server: A gateway server
is used to provide connectivity to Communication Equipment
other networks. The GIAS (Gateway
• Repeater: A repeater operates
server of VSNL, India) is a gateway
at layer 1. It has just enough
server that provides connectivity to
intelligence to determine the
all other networks on the Internet
incoming signals at layer 1 and then
for the Indian segment of the
sends out a clean stream of signals
Internet. A gateway server also
built from scratch, eliminating
provides connectivity to dissimilar
noise. A repeater has one incoming
networks.
and one outgoing line, extending
• Communication server: the distance a signal may be sent
Communication servers are more over a transmission media.
diverse than other servers. The main
• Hub: A hub is a multi-port repeater.
functions of a communication server
Any incoming signal is repeated
are linking client workstations on
on all other outgoing lines. A hub
the LAN with mainframe computers,
functions at layer 1.
sharing a pool of modems among the
client stations, and communicating • Bridge: A bridge has more
with other LANs as well. Examples intelligence than a hub or
of the most popular LAN servers repeater. This device separates
include Novell Netware LAN Server, two segments of a single LAN.
Microsoft Windows NT Server or A bridge operates at layer 2 by
Windows 2000, IBM OS2 Server, etc. examining the destination address
in the frame header. By consulting
• Workstations: The clients that
a table, the bridge determines if
utilize the network server’s
the frame needs to pass on to the
services are workstations, also
other segment. Only certain frames
known as nodes. Special software
may pass, those with the correct
is installed on workstations so they
MAC address.
can communicate with the server
and use the services. Examples • Switch: A switch is a multi-port
 LOCAL AREA NETWORK/WIDE AREA NETWORK 105
bridge that operates at layer 2. It looks at MAC layer
addresses like a bridge, consults a table, and determines if a
frame needs to be sent on one of the attached lines. Multiple
connections can occur through a switch simultaneously as
long as they do not compete for the same line. A switch,
like a bridge, does not divide a LAN into two LANs but
reduces unnecessary traffic on LAN segments.
• Router A router is more intelligent than bridges and switches.
It functions at layer-3. At layer-2 all incoming frames are
checked for errors and then they are delivered to layer-3.
A layer-3 datagram has an address that permits it to be
sent across internetworks (interconnected networks). This
requires a single global addressing scheme. The router
finds the layer-3 address and consults a table that it keeps.
There the router will learn which attached line to send the
datagram on. A router provides facilities to the stations on
the LAN to access other networks.
Remember
A LAN comprises
3.1.5. Devices cables, access
A LAN device is essentially any device directly connected to a points, switches,
routers, and
local area network. Through a shared access point like a router, other components
devices on the same local area network can usually share files, that enable
access resources, and connect to the Internet. To ensure they stay devices to
cool and secure from tampering, many LAN devices we use daily connect to
internal servers,
are stored in protected server rooms. The definition of a LAN has
web servers,
remained unchanged for over 20 years, even as the devices within and other LANs
a LAN continue to evolve. A local area network, or LAN, is a small via wide area
data network that can be set up in a home or office, enabling networks.
devices like PCs to share services and resources. Ethernet and
Wi-Fi (802.11x) are the most popular LAN technologies nowadays,
with older technologies like Token Ring rarely used. Essentially, any
device in your home, workplace, or classroom directly linked to a
local area network via Ethernet cable or Wi-Fi adapter is considered
a LAN device. When two devices are connected through the same
access point, such as a router, they can typically share files, access
resources, and use the Internet. Many LAN devices we use daily
are kept secure and cool in server rooms.

3.1.5.1. LAN Devices You See Daily

LAN devices include the router and modem found in the majority
of homes and small offices, as well as computers, tablets, and
smartphones that are linked to a network. If your smart appliances
are networked, they can also function as local area networks (LANs).
These can range from IP-based webcams and network printers
106 COMPUTER NETWORKS

to Blu-ray players, smart TVs, and game 3.1.6.1. Switches

consoles. These days, some manufacturers
are producing smart ovens, dryers, and The main piece of hardware in LANs that
washers that, once set up for your home allows network communication is a switch.
network, can also function as LAN devices. Cabling is run to every hardware device
Any device that is directly connected to a on the network in order to create a LAN.
local area network can be considered a LAN Switches are used to allow data flow and
device in the most basic sense. When two connect hardware cabling throughout a local
devices are connected by the same access area network. Every switch has ports built
point—such as a router—they can typically in for network connections; the quantity of
share files, access the same resources, and ports depends on the switch’s size. Switches
use the Internet. Numerous LAN devices have features like quality of service to
that you might use on a daily basis are prioritize different kinds of LAN traffic and
kept cool and safe from tampering by being redundant power supplies that allow them
kept in secure server rooms. to operate around the clock, seven days a
week.

3.1.5.2. LAN Devices in a Large Network

3.1.6.2. Wireless Access Points
Large corporations’ and educational
institutions’ networks often contain extra Wireless access points can be added to
network equipment that isn’t typically the LAN to increase flexibility and lessen
found in a home. Among them are servers, the need for cabling. This hardware device
which are powerful PCs that store data and allows for workplace mobility by transmitting
applications and control network access using radio waves to computers through antennas.
stored passwords. Advanced networking Common locations for installations include
equipment like switches, firewalls, and conference rooms and open spaces like
routers are also utilized in large networks. cafeterias. To expand network services, the
These networking devices manage traffic access point is connected to a LAN switch.
and direct data to the appropriate devices, It’s crucial that this device has security
similar to a home router. The key distinction features, such as data encryption, when
is their sophisticated management software connecting it to a LAN.
and ability to handle numerous connections
simultaneously. 3.1.6.3. Servers
Servers are business-class PCs that provide
3.1.6. Hardware Devices for collaboration tools, antivirus protection,
LAN messaging, and other services to the office.
Similar to switches, servers run around the
When two or more PCs are connected, a
clock, seven days a week, to satisfy the
local area network, or LAN, is established.
demands of the office. Multiple processors,
The hardware added to a local area network
larger memory, and power supplies are
(LAN) as it expands can include routers,
features found in servers. To ensure uptime,
wireless access points, servers, switches
server components can be changed while
for connecting hardware (such as computers
the machine is operating. Servers are housed
and printers), and wireless access points.
in secure data centers and computer rooms
and are built for high availability.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 107
3.1.6.4. Thin Client • They are easy to install and
maintain.
Installing desktop computers is an option
• hey exist as connected (wired) and
provided by thin client hardware. Thin
wireless configurations.
clients are hardware devices with a smaller
profile and fewer components. These LAN can provide the following services.
devices consume less power, leading to They are:
reduced operating costs. They establish
• File-based services. File transfer
a LAN connection and provide ports for
between nodes inside a local
a keyboard, mouse, and monitor. Workers
area network (LAN). When a new
rely on servers to access application
transaction is made on behalf of
services through thin clients. Unlike
a customer, for instance, the file
desktop computers with local file storage
containing the specific customer’s
capabilities, thin clients depend on storage
detailed transactions is transferred
connected to the LAN. This centralized
from the server to the client in a
approach to providing application services
typical local area network (LAN)
and file storage simplifies management and
used for local banking. Additionally,
enhances security.
it offers effective means of data
retrieval and storage. The presence
3.1.7. LAN Advantages and of multiple copies of an identical
file facilitates the essential
Services synchronization required for file
A local area network (LAN) consists of updates. In order to enable safe
personal computers (PCs) that connect and recovery in the event of a failure,
share resources like printers and storage via LANs additionally offer backup for
phone lines or cables. LANs are typically vital data. Additionally, it offers the
confined to a single building or a small ability to encrypt data so that only
area, but this is not a strict rule. The area specific individuals may access it.
could be a home or workplace with only • Print services. Applications may use
a few devices, or it could span a larger one or more printers, modems, and fax
space like an entire office building with machines that are connected to a local
numerous devices. The common feature of area network (LAN). These devices
all LANs, regardless of size, is connecting can be shared by several clients. Many
devices within a limited area. The benefits clients can safely share printers and
of networking devices within a LAN include other similar devices thanks to print
file sharing, printing to shared printers, services. These services are available
using a single internet connection, and on all modem LANs.
more. Larger LANs may include dedicated
servers for email, global user directories, • Application-based services. Higher
and other shared corporate resources. processing power may be needed
for client-side applications. The
A LAN has the following advantages: power of servers exceeds that of
clients. A server’s processing power
• They work on higher operating can be shared by multiple clients
speed than WAN and MAN. over a local area network (LAN).
• They suit the requirements of a A good illustration of this is with
specific organization. application servers.
108 COMPUTER NETWORKS

• Mail and message-based services. One useful tool for

information sharing on a network is electronic mail (e-mail).
Emails can be sent to another user or stored. On the Internet,
there are a lot of free mail servers that offer their users
free mail services.
• Database services. Storing and retrieving data in databases
is another requirement of a client to control and manipulate
Keyword the data. Database servers are the most popular ones that
service the need of clients.
Database server
is a server that • Distributed data services. When data is distributed, it
uses a database is possible for more than one client system to share the
application to data. Though the databases appear physically distributed,
provide database
services to there is a single logical view given by the database server.
other computer There are many issues involved in sharing a single file by
programs or more than one client. Consistency issues are very seriously
computers, as tackled when updating shared files.
defined by the
client–server • Remote services. Linking a LAN with a remote computer
model. or a mainframe is another required service.
This feature is used to access remote databases available on
large mainframes.

3.1.8. Protocol
A local area network is one that operates within a structure, such
as a home office, business, or office. LANs are utilized in confined
geographic areas. Within a constrained network distance, LANs can
transfer data at speeds of up to 10Mbps. Peer-to-peer networking and
client-server networking are the two forms of network architecture
that can be used in local area networks. Ethernet and Token Ring
connections are the most often used types in local area networks
(LANs).
The ability of LAN protocols to effectively transfer data over
shorter distances—a few hundred feet, for example—through a
variety of media—including copper cabling—sets them apart. There
are various “layers” of the “Open Systems Interconnect,” or OSI,
model, and distinct protocols for various uses. Generally speaking,
lower level, or physical, layers are meant to be described when a
protocol is referred to as “LAN.” The most widely used LAN protocols
are “Fiber Distributed Data Interface,” or “FDDI,” “Ethernet,” and
“Token Ring.” By far the most popular kind of LAN protocol is
“Ethernet.” It is identifiable by its typical “CAT5” copper cable
medium and is found in homes and offices all over the world. To
exchange data, all systems connect to a switch or hub.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 109
The outdated LAN technology known or a constrained geographic area. Both
as “Token Ring” is becoming less common. wired and wireless connections between
The fundamental idea behind “Token Ring” the devices are possible. Standard LAN
is that only the intended recipient can read technologies include IEEE 802.11 Wireless
a single “token” that is transferred between LAN, Ethernet, and Token Ring.
systems or via a hub. The term “FDDI”
describes the fiber cabling method used to • Ethernet: The most popular LAN
transfer LAN traffic. When greater security technology, as defined by IEEE
is needed or longer distances—typically standards 802.3, is Ethernet.
between floors or buildings—are needed, Ethernet’s widespread use is due
fiber cabling is utilized. Multiple network to its simplicity in understanding,
protocols can be used by LANs (Local Area implementation, and maintenance,
Networks) to define their networks. The OSI as well as its ability to enable
Data Link Layer defines these protocols. low-cost network implementation.
Additionally, Ethernet allows for
• ARP (Address Resolution flexibility in terms of permitted
Protocol): Converts IP address to topologies. The Physical Layer and
MAC Address. An ARP cache is Data Link Layer are the two OSI
maintained in the system that has model layers in which Ethernet
a table mapping of IP address to functions. Since we primarily work
MAC address in the Network. with DLL, the protocol data unit
• RARP (Revers Address Resolution for Ethernet is the frame. Ethernet
Protocol): Convers MAC address to uses CSMA/CD as its access control
IP Address mechanism to handle collisions.
Ethernet employs the Manchester
Classifications of LAN transmissions:
Encoding Technique.
• Unicast: Data is transmitted from Since we are talking about IEEE 802.3
source to a destination system standard Ethernet, therefore, 0 is expressed
• Multicast: Data is transmitted from by a high-to-low transition, a 1 by the
source to multiple destinations low-to-high transition. In both Manchester
using a special multicast IP address Encoding and Differential Manchester,
Encoding Baud rate is double of bit rate.
Baud rate = 2* Bit rate
Did you know?
Ethernet LANs consist of network nodes
The development and proliferation of personal
and interconnecting media or link. The
computers using the CP/M operating system in the
late 1970s, and later DOS-based systems starting in network nodes can be of two types:
1981, meant that many sites grew to dozens or even
hundreds of computers. • Data Terminal Equipment (DTE):
DTEs are typically the final devices
that change signals received into
user information or vice versa.
3.1.9. LAN Standards Personal computers, workstations,
A local area network, or LAN, is a type file servers, print servers, and
of data communication network that links other devices commonly referred
computers or terminals within a building to as end stations are examples
110 COMPUTER NETWORKS

of DTEs. These devices are either on various Hawaiian Islands. At the

data frames’ source or destination. data link layer, Aloha is a multiple
The data terminal equipment can access protocol that suggests a way
consist of one unit or several units for multiple terminals to access the
that are networked together and medium without causing a collision
fulfill all the necessary tasks to or interference.
enable user communication. DTE There are two different versions of
and users can communicate with ALOHA:
each other.
• Data Communication Equipment 1. Pure Aloha: Pure Aloha is a
(DCE): The intermediary network decentralized, easy-to-implement
devices, or DCEs, are responsible protocol that is not slotted. The
for receiving and forwarding frames stations in Pure Aloha simply send
over the network. They could be frames whenever they have data
stand-alone gadgets like routers, to transmit. Prior to transmission,
network switches, and repeaters, they do not verify if the channel is
or they could be communications busy. In case two or more stations
interface units like modems and transmit simultaneously, collisions
interface cards. The DCE may occur, leading to the destruction
be a component of the DTE or of frames. Upon sending a frame,
intermediate equipment and carries each station expects to receive an
out tasks like coding and signal acknowledgment from the recipient.
conversion. If the acknowledgment or frame is
not received within a specified time
Currently, these data rates are defined frame, the station assumes it has
for operation over optical fibers and twisted- been lost and retransmits the frame
pair cables: after a certain waiting period.
i) Fast Ethernet. Fast Ethernet refers This unpredictability helps in
to an Ethernet network that can reducing collisions. While this
transfer data at a rate of 100 Mbit/s. approach works well in small
ii) Gigabit Ethernet. Gigabit Ethernet networks with light traffic, it
delivers a data rate of 1,000 Mbit/s performs poorly in heavily loaded
(1 Gbit/s). networks. This limitation led to the
development of Slotted Aloha. To
iii) 10 Gigabit Ethernet. 10 Gigabit
ensure the effectiveness of Pure
Ethernet is the recent generation
Aloha, certain conditions must be
and delivers a data rate of 10 Gbit/s
met:
(10,000 Mbit/s). It is generally
used for backbones in high-end 1) All frames should be of the same
applications requiring high data length.
rates. 2) Stations cannot generate frames
• ALOHA. The University of Hawaii while sending or attempting to
project included the design of the send frames.
Aloha protocol. It allowed packet 3) The population of stations uses a
radio networks to be used to Poisson distribution to determine
transfer data between computers how to transmit, including new
 LOCAL AREA NETWORK/WIDE AREA NETWORK 111
frames and previously collided 2. Slotted Aloha: The only real
frames. difference between this and Pure
Vulnerable Time = 2 * Tt Aloha is how transmissions happen.
Rather than sending the message
Efficiency of Pure ALOHA: at the exact moment of demand,
the sender holds off. The shared
Spure= G * e^-2G
channel’s time is split into distinct
where G is number of stations wants intervals known as Slots in slotted
to transmit in Tt slot. ALOHA. Only at the start of the slot
Maximum Efficiency:Maximum are the stations allowed to send a
Efficiency will be obtained when frame, and each slot can only have
G=1/2 one frame sent. A station must wait
(Spure)max = 1/2 * e^-1 = 0.184 until the start of the following time
slot if it is unable to get the frame
which means, in Pure ALOHA, only onto the channel at the start of the
about 18.4% of the time is used for slot. If two stations attempt to send
successful transmissions. at the start of the same time slot,
there is still a chance of a collision.
However, compared to Pure Aloha,
the number of collisions that could
occur is still significantly decreased,
and performance is significantly
improved.
Collision is possible for only the
current slot. Therefore, Vulnerable
Time is Tt.
Efficiency of Slotted ALOHA:
Sslotted = G * e^-G
Maximum Efficiency:
Figure 3.2. Throughput and rate of transmission
(Sslotted)max = 1 * e^-1 = 1/e = 0.368
of the frame (Source: Geeksforgeeks, https://
media.geeksforgeeks.org/wp-content/ Maximum Efficiency, in Slotted ALOHA,
uploads/20230710122406/ETHERNET_2.webp). is 36.8%.

Did you get it?

1. What are the different types of devices commonly found in LAN environments?
2. What hardware components and infrastructure are required to establish and maintain
a LAN?
112 COMPUTER NETWORKS

3.2. WIDE AREA NETWORK

→Learning Objectives

• Identify common WAN protocols and their characteristics.

• Describe the functions and features of WAN devices.
• Learn how to connect WAN devices to establish a wide area network.
• Explore strategies for managing and optimizing wide area networks.
• Describe various types of WAN connections available in networking.
• Identify key WAN standards and their significance in networking infrastructure.
• Explore the configuration and usage of WAN ports in wide area network setups.
• Acquire the skills necessary to design and implement a wide area network.

Wide area networks, or WANs for short, are communication networks that connect
local computer networks to a wider, functional network that can span both domestic
and foreign locations by utilizing current technology. On the other hand, the local area
network and the metropolitan area network offer communication inside a constrained
geographic area. It’s often desirable to link two computer networks together, especially
for companies with multiple locations. Using current telephony technology makes this
process easier, starting with the local area network and moving on to the wide area
network. Fiber optics are commonly used to establish connections between networks
in different facilities. This typically involves using PSTN (public switched telephone
network) technology or POTS (standard phone lines). Another option for network creation
emerged in the 1990s: ISDN (integrated services digital network) solutions. This was
mainly because it became more cost-effective to expand the network across international
borders.
A wide area network, or WAN, allows businesses to use shared resources to operate.
For example, many retail pharmacies use it to help customers who fill prescriptions at
their stores. Once a customer is in the pharmacy’s common customer database, they can
fill a prescription at any of the company’s locations—even when they’re on vacation in a
different state. Businesses also benefit from using wide area networks. With this kind of
wide area network application, internal operations like sales, marketing, production and
development, and accounting can also be shared with approved locations. Employees
can work from almost anywhere by using this to combine disparate location-based
 LOCAL AREA NETWORK/WIDE AREA NETWORK 113
computer networks into a single computer information,
network for the entire company. In the event • Sets up a connection to the far-end
that a natural disaster damages or renders servers,
one facility unusable, staff members simply
• Sends the information to the far-end
relocate and continue working using the
servers,
shared network at their new location.
• Receives an acknowledgment
The current GPRS, EDGE, or 3G mobile from the far-end servers with the
network is expected to be used for the Wide message sequence number of the
Area Network because it is widely accessible last information received,
and has coverage in the majority of the UK.
• Completes Cyclic Redundancy
The crucial aspect in this situation is that
Checks (CRC) successfully, and
data from the Personal Connector must be
routinely sent to the distant servers for • Sends a ‘Success’ message back to
examination. Installing a software program the Personal Connector with the
on a mobile device allows it to interact message sequence number and
with both the server application and the receives an acknowledgment, sent
Personal Connector. This software, known by the Personal Connector.
as service integration software, is essential • All pointers and information are now
for facilitating the communications needed updated. A timeout occurs, and the
for the analysis of data on smart clothing. data is resent in failure scenarios,
such as when acknowledgments are
There may be times when the mobile not received or CRC malfunctions.
network doesn’t receive any signal, Whether it is GPRS or 3G, this
preventing data from being sent to distant software program will provide the
servers. In such situations, data is retrieved interface to the selected mobile
from the Personal Connector and stored on network. The software program
the mobile device. The Personal Connector can be updated in the future in
holds onto this data until the end servers response to changes in the wide
can be reached and the data transfer is area network, which could involve
completed successfully. Each piece of data Wi-Max or any other technology
is time-stamped and assigned a message selected for 4G mobile. Establishing
sequence number. This two-stage process a reliable protocol is essential for
enhances the overall service reliability, the complete data transfer process
ensuring that information can be transmitted from the home network to the
to the far-end servers once the wide area distant servers through a wireless
mobile network is accessible. This eliminates wide area network.
the need to wait for data transfer from the
Personal Connector to be successful.
3.2.1. WAN Protocol
In a normal scenario,
Packet-switching exchange nodes (PSEs)
• This application polls the Personal are the hardware that drops traffic onto
Connector, the wires connecting sites in standard-
sized packets, delivered in order, with error
• Retrieves the information,
correction. X.25 is one of the oldest protocols
• Checks the timestamps and used to deliver WAN traffic. Leased lines,
message sequence numbers of this dial-up phone services, and Integrated
114 COMPUTER NETWORKS

Services Digital Network (ISDN) connections are examples of the

physical links. It’s not as common these days.
The replacement for X.25 is frame relay. Frame Relay divides
data into frames of varying sizes and leaves it up to the endpoints
to handle packet loss and error correction. The overall data rate is
accelerated by these discrepancies. Furthermore, Frame Relay uses
fewer dedicated connections to build mesh networks, resulting in
fewer physical circuits and a financial savings for businesses. Once
more, frame relay was once very popular but is now less so. The main
distinction between frame relay and asynchronous transfer mode
(ATM) is that data is divided into uniformly sized packets known
Keyword as cells. Cells facilitate the blending of various traffic classes onto
Integrated a single physical circuit and increase the ease with which service
Services Digital quality can be guaranteed. ATM’s drawback is that, due to its use
Network is a set of relatively small cells, a disproportionately large portion of each
of communication
standards for cell’s contents is taken up by the headers. As a result, frame relay
simultaneous uses bandwidth more efficiently overall than ATM. Additionally,
digital transmission ATMs are becoming less popular with business clients.
of voice, video,
data, and other
network services
Today, a lot of corporate data is transported across WAN links
over the digitalized using multi-protocol label switching. MPLS routers in an MPLS
circuits of the network can swiftly determine where to forward packets and assign
public switched
telephone network.
them to the appropriate class of service based on short header
segments known as labels. As traffic moves between sites, this
enables the operation of various protocols inside MPLS packets
while allocating the proper priority to various applications. One
protocol that is frequently carried within MPLS is Internet Protocol
(IP), which became more widespread in the 1990s. The Wide Area
Network, or Internet, rose to prominence recently because it is
so widely used and makes business management and information
sharing possible everywhere in the world. The operation of this
complex WAN structure is governed by a set of rules and standards,
just like any other type of activity. The technologies or protocols
known as WAN protocols control how effectively WANs operate.
These significantly affect business continuity, which is important
for enterprises.
Depending on the features they support and the protocols
governing those features, technological appliances can operate
effectively. Wide Area Networks, or WANs, employ a variety of
networking tools and technologies. They are not the same as what
Local Area Networks (LANs) use. The majority of WAN technologies
and protocols are Layer 2 (data link layer) protocols. Asynchronous
Transfer Mode (ATM), Frame Relay, Point-to-Point Protocol (PPP),
Synchronous Optical Network (SONET), Synchronous Digital Hierarchy
(SDH), and X are the main WAN protocols that are currently in use,
 LOCAL AREA NETWORK/WIDE AREA NETWORK 115
as well as a few additional WAN protocols. Internet access is referred to as
Gaining a deeper comprehension of these broadband, or broadband Internet
ideas is crucial because they provide access, and because of its high
the foundation for improving Internet data transmission rate, it is one
applications, which in turn helps businesses of the most widely used WAN
compete in the market. A quick summary protocols. Through broadband,
of some of the most significant protocols speeds of up to 256 kbit/s are
is provided below. possible. In most places, cable
modems and digital subscriber line
• Asynchronous Transfer Mode: The (DSL) are the standard broadband
packet switching protocol known technologies. Fiber-optic cables
as ATM allows data traffic to be and Very High Bitrate DSL (VDSL
encoded into discrete, pre-sized or VHDSL) are two examples of
cells. The foundation of this protocol more recent technologies that have
is connection-oriented technology. developed. Modem technology
It functions by creating a virtual is the foundation of DSL, which
circuit between two endpoints allows Internet access over copper
even before any data is exchanged. phone lines. A Digital Subscriber
A mechanism for routing traffic Line Access Multiplexer is used
on a data stream with fixed 53- to connect multiple DSL users to
byte packets is established by the high-speed network (DSLAM).
the ATM protocol suite. Fixed- The telephone company connects
sized cells reduce the need for available DSL lines to the ATM
switching and multiplexing. Two network by using the DSLAM to
systems on the network must notify aggregate data transmission from
each intermediate switch about those lines. After demultiplexing
service requirements and traffic the signals at the transmission
parameters for communication endpoint, a DSLAM forwards the
to be established. ATM operates data to a DSL connection.
over layer 1 links to provide data
link layer services as a cell relay The various DSL connection types—
protocol. ATM’s inherent circuit- ADSL, HDSL, SDSL, IDSL, and VDSL—
switched and packet-switched collectively called xDSL are what create
network properties make it ideal for the connection between the phone
real-time media transport and wide company and the office. Discrete Multitone
area network data networking. ATM Technology (DMT), Simple Line Code (SLC),
Adaptation Layers (AALs) enable a Carrierless Amplitude Modulation (CAP),
variety of services. Between the Multiple Virtual Line (MVL), and Discrete
ATM layer and an upper layer, Wavelet Multitone (DWMT) are some of
the AAL transfers the ATM cells. the modulation technologies used in Digital
It carries out a crucial task known Subscriber Lines. The foundational network
as Segmentation and Reassembly system of the early WAN protocols, the
(SAR). While information is being Integrated Services Digital Network (ISDN),
received from the higher level, data offers digital phone connectivity and permits
is divided into ATM cells. high-speed simultaneous transmission
of data, voice, video, and graphics over
• Broadband Access: High-speed standard communication lines using bearer
116 COMPUTER NETWORKS

channels. In addition to telephone networks, the Internet via dial-up. Connection

packet-switched networks can also be used establishment over synchronous and
to transmit ISDN. Primary Rate Interface asynchronous circuits is facilitated
(PRI) and Basic Rate Interface (BRI) are by this protocol. Encapsulation is a
the two fundamental types of ISDN service. crucial component of PPP, whereby
Large enterprises and other users who need PPP frames are encapsulated in
more capacity can use PRI, which offers a lower layer protocol to provide
a total of 1536 kb/s. BRI is limited to 160 framing and other functions like
kb/s for individual users. transmission error detection. The
two common encapsulated PPP
• Frame Relay. Frame Relay is forms are Point-to-Point Protocol
a WAN protocol that is only over ATM (PPPoA) and Point-
for internetworking Local Area to-Point Protocol over Ethernet
Networks (LANs), or the transfer (PPPoE). They are used to carry
of data between LANs and WAN out tasks similar to those that DSL
endpoints. It is based on packet- services complete.
switching technology. Network
providers frequently use frame • SONET/SDH. Light Emitting
relay, a low-cost technique, Diodes (LEDs) or lasers are used
to encapsulate data and voice in the multiplexing WAN protocols
between LANs connected by a Synchronous Optical Networking
wide area network. A private or (SONET) and Synchronous Digital
leased line will connect the user Hierarchy (SDH) to facilitate the
to a frame relay node. Frame relay transport of multiple digital bit
enables data transfer between user streams over the same optical fiber.
devices across numerous switches SDH and SONET are two closely
and routers at both the physical related protocols that use circuit
and data link layers. The amount mode communication as their
of frame relay technology being foundation. Instead of using copper
used has been steadily declining cables, SDH tests the compatibility of
as a result of advancements in optical-fiber-based digital telephone
other technologies. However, this links. When widely implemented,
affordable WAN protocol with SONET/SDH allows multiple ISPs
continuous connectivity at 64 kb/s to concurrently share an optical
is still in use in rural areas that fiber without interfering with each
have not yet had access to DSL or other’s traffic load. Time division
cable modem. multiplexing (TDM) protocols
are physical layer protocols that
• Point-to-Point Protocol. A direct provide uninterrupted connections
connection between two nodes can without requiring packet mode
be established via serial cables, communication.
fiber optic links, exclusive radio
links, cellular phones, telephone • X.25: X.25 is a packet-switched
lines, and trunk lines using the network-based WAN protocol for
PPP data link protocol. Customers WAN communications. It delineates
frequently use point-to-point data exchange and control of
protocol, a WAN protocol, to access information within a user appliance,
 LOCAL AREA NETWORK/WIDE AREA NETWORK 117
Data Terminal Equipment (DTE), apart, it will be necessary to connect these
and a network node, Data Circuit locations in order to share, exchange, and
Terminating Equipment (DCE). X.25 manage data or communication. In order
comprises physical links such as to accomplish this, the company requires a
packet-switching exchange (PSE) Telecommunication Service Provider (TSP) to
nodes for networking hardware, establish connections between the local area
leased lines, and telephone or networks at each location. Long-distance,
ISDN connections. Its unique expansive area networks are overseen by
functionality is its capacity to work telecommunications service providers. Voice
effectively on any type of system and data communications are carried over
that is connected to the network. different networks by TSPs. Wide Area
X.25, although replaced by superior Networks (WANs) are the networks that
technology, continues to be in use. link Local Area Networks (LANs) in distinct
It utilizes a connection-oriented geographic locations.
service that enables data packets
to be transmitted in an orderly WANs’ Main Function:
manner. 1. WANs generally connect devices
• WAN solutions to address specific that are separated by a broader
needs. To maintain its wide area geographical area than can be
network (WAN) infrastructure, served by a LAN.
which is the foundation of 2. WANs use the services of carriers,
communication, any business needs such as telephone companies, cable
improved technology solutions. companies, satellite systems, and
Technological innovators provide a network providers.
full range of WAN solutions that are
3. WANs use serial connections of
specifically designed to fortify an
various types to provide access to
organization’s WAN infrastructure.
bandwidth over large geographic
Advanced features offered by
areas.
these technology products
include data compression, VPN
security, encryption capabilities, 3.2.2.1. Modems
WAN acceleration, optimization, When transmitting and receiving
dependability, security, and information, modems allow digital data to
dynamic load balancing of incoming be sent over analog media. A voice band
and outgoing data traffic. These modem translates digital signals generated
solutions are essential to the by a computer, which are 1s and 0s, into
competitive business world as voice frequencies that can be sent over
they determine whether wide area the telephone network’s analog lines.
networks remain stable and help To be input into a computer or network
organizations meet their business connection, the sounds are converted back
continuity objectives. into a digital signal on the other end of the
connection by another modem.
3.2.2. WAN Devices
In situations where a company or organization
has locations that are geographically far
118 COMPUTER NETWORKS

3.2.2.2. CSU/DSU
A combined piece of equipment called a Channel Service Unit (CSU)
and Data Service Unit (DSU) is used to monitor line clocking and
frame synchronization. It is also capable of error detection at the
physical layer; in a way, it functions like a modem.
Remember
Wide-area
3.2.2.3. Access Server
network (WAN)
optimization is
Concentrates dial-in and dial-out user communications. An access
a collection of server may have a mixture of analog and digital interfaces and
techniques that support hundreds of simultaneous users.
improve WAN
performance
metrics such 3.2.2.4. WAN Switch
as throughput,
congestion, A multi-port internetworking device used in carrier networks. These
and latency. devices typically switch traffic such as ATM and operate at the Data
WAN design,
Link layer of the OSI reference model. Public switched telephone
technology
choices, and network switches may also be used within the cloud for circuit-
traffic flow switched connections like Integrated Services Digital Network (ISDN)
arrangements or analog dial-up.
all affect WAN
performance.
3.2.2.5. Router
A router provides internetworking between the LANs and WAN
access interface ports that are used to connect to the service provider
network. These interfaces may be serial connections or other WAN
interfaces. With some types of WAN interfaces, an external device
such as a DSU/CSU or modem (Analogue, Cable, or DSL) is required
to connect the router to the local point of presence (POP) of the
service provider.

Core Router
A router that resides within the middle or backbone of the WAN
rather than at its periphery. To fulfill this role, a router must be able
to support multiple telecommunications interfaces of the highest
speed in use in the WAN core, and it must have the ability to forward
IP packets at full speed on all of those interfaces. The router must
also support the routing protocols being used in the core.

3.2.3. Connecting Your Wide Area Network (WAN)

Devices
The connection between your router and the service provider must
be established before you can utilize any type of WAN connection.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 119
The type of connection will dictate your next DCE. Essentially, one end of the
steps. Commonly used interfaces include cable that connects to your router
ISDN ports, serial ports, or Channel Service links to the DTE, while the end
Units/Data Service Units (CSU/DSU). connecting to the service provider’s
equipment links to the DCE.
• Serial ports. A serial port is a
primary method to connect the • Integrated CSU/DSU. Integration
router to your service provider’s of the CSU/DSU components into
network. Depending on your service routers has grown in popularity
provider, you may be provided with as a way to cut costs or simplify
a CSU/DSU. The CSU/DSU readies systems. You can have the CSU/
your digital data for transmission DSU functionality built into your
over a telephone company network router’s circuitry, saving you from
that may not support digital signals, having to place another large
similar to a modem (modulator/ device in your communications
demodulator). These devices room. The disadvantage is that you
connect to the T1 link or service might have to replace your router
provider connection, which in turn if you choose to switch the kind of
links to your router. connection you have. As a result,
Cisco developed the WAN Interface
Data Terminal Equipment (DTE) Card (WIC) port, which enables you
refers to end-user devices, while to reuse your current router and
Data Communications Equipment switch at a minimal cost the type
(DCE) refers to telephone of service provider connection you
company devices, categorizing use.
the connection into two device
types. In your interactions with • ISDN ports. Instead of using WIC
the service provider, your router ports that are already set up for
acts as the DTE, and the service ISDN environments, Cisco has
provider’s communication devices produced and marketed routers
act as the DCE. In the realm of with a fixed configuration due to
computing, a computer serves as the increasing popularity of ISDN
the DTE, and a modem as the DCE. connections. Although you have
To test your network equipment few options for connectivity, if you
before deployment, you can utilize a don’t plan on switching up the
specialized serial cable known as a connection types you use with your
back-to-back serial cable, designed service provider, the cost of these
with distinct ends for DTE and routers may be appealing.
120 COMPUTER NETWORKS

Q: What is the key architectural components of WAN?

A: The exact design of your WAN architecture will vary based on your business
requirements and the type(s) of WAN in use. However, seven architectural components
are essential for WAN implementation:
1. End devices. End devices are computing systems that are being connected through
wide area network technology.
2. Customer premises equipment (CPE). While most of the networking infrastructure
used in WAN is owned by the carrier or service provider, CPE is owned by the
enterprise and sits in your local environment.
3. Access points and routers. While access points and routers technically fall under
CPE (as they are located within the LAN environment), they deserve special mention
as indispensable architectural components of WAN. A router is your core LAN hub
and manages data flow to and from your end devices.
4. Network switches. Network switches (also part of your CPE) sit between your
access points and routers and the end devices to which they are connected.
5. Local area network (LAN). LAN is a key architectural component of a wide area
network. All the components we discussed so far (end devices, CPE, access points,
routers, and network switches) are part of your local area network environment.
6. Connecting media. The connecting media carries data packets in a WAN mesh
architecture to and from different LAN components. There is a wide variety of media
to choose from when setting up your WAN, including traditional MPLS lines, fiber
optic cables, the cellular (3G, 4G, or 5G) spectrum, and satellite connectivity.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 121
7. Metropolitan area network (MAN) – optional. A metropolitan area network or
MAN is an optional component of WAN architecture. Instead of directly connecting
two or more LAN environments, you can choose to connect LANs within a relatively
small radius using a MAN.
8. Unified management portal – optional, but recommended. A unified WAN management
portal is becoming increasingly commonplace in enterprise WAN implementations. This is
because WAN can be difficult to orchestrate and manage, with each LAN hub responsible
for network management for its own location.

3.2.4. WAN Management and duplication (short for reducing redundant

data).
Optimization
A wide area network (WAN) is a network that The ability to shape traffic allows for
connects smaller campus and metropolitan the prioritization of certain applications (like
networks spread across different locations VoIP) over less urgent ones (like email),
into a single, distributed network by means thereby enhancing WAN performance
of a variety of links, including private lines, overall. This can be codified into quality of
virtual private networks (VPNs), wireless service settings that identify traffic classes
(cellular), MPLS, and the Internet. The based on the bandwidth allotted to each
locations they link may be a few miles class, the kind of WAN connection each
apart or located halfway across the world. traffic type will use, and the priority each
A wide area network (WAN) can be used class receives in relation to the others.
in an organization to share corporate
communications and resources by linking 3.2.4.1. WAN Optimization
branch offices and even individual remote
workers to the data center or headquarters. Enterprise WAN performance problems are
frequently caused by latency and bandwidth
The greater the distance between limitations. To combat them, WAN
two devices, the longer it will take for optimization appliances employ a number
data to travel between them because of strategies, such as traffic shaping, local
data transmission is still subject to caching, protocol optimization, deduplication,
the laws of physics. There is a direct and compression. With the use of less
correlation between distance and delay. expensive bandwidth connections, typically
Performance issues can also be brought in the form of commercial internet services,
on by network congestion and dropped along with traffic shaping and quality of
packets. By improving the efficiency of service tools -- to increase reliability -- SD-
data transmissions, WAN optimization can WAN CPE or platforms offer an additional
help with part of this. This is significant degree of application performance control.
because wide area networks (WANs) can
be costly. Consequently, technologies have
emerged to minimize traffic over WAN links 3.2.4.2. WAN Security
and guarantee efficient delivery of data.
Whenever end users will be using their
These optimization techniques include
devices, including those who work from
compression, caching (moving frequently
home, the security of the WAN should be
used data closer to the end user), and de-
extended. Firewalls and antivirus programs
122 COMPUTER NETWORKS

should be used by end users who use WAN to guard against device
compromises and illegal access. In addition to providing WAN
connectivity, using a VPN has the added advantage of encrypting
Keyword data. Network devices connected to a WAN from a remote location
Virtual private should also require users to establish a VPN connection. Furthermore,
network (VPN) a key-exchange feature of SD-WAN is utilized for device authentication
is a mechanism for
creating a secure across various endpoints.
connection between
a computing It is not reasonable to expect a certain level of security from a
device and a WAN service provider, even though a WAN can be just as secure. A
computer network, WAN system’s complete security is not guaranteed, not even with
or between two
networks, using a VPN in place. A hacker previously obtained access to Microsoft
an insecure by breaking into a user’s home device, which then followed the
communication VPN back to Microsoft.
medium such as the
public Internet.

3.2.5. Types of WAN Connections and how WAN

Connections Work
Wired and wireless technologies are both possible for WAN
connections. T1s, carrier Ethernet, multiprotocol label switching,
and commercial broadband internet links are examples of wired WAN
services. Cellular data networks like 4G LTE as well as public WiFi
and satellite networks are examples of wireless WAN technologies.
Although most businesses still prefer wired network connections
over WANs, wireless WAN technologies based on the 4G LTE
standard are becoming more and more popular. WAN infrastructure
can be either privately held or leased as a service from a third-
party provider, like a cable company, internet service provider,
telecommunications carrier, or private IP network operator. The
service itself can run on a shared, public channel like the internet
or a dedicated, private connection that is frequently supported by
a service-level agreement. A mix of private and public network
services is used by hybrid wide area networks. Software-defined
WAN (SD-WAN) is designed to make hybrid WAN architectures
easier for enterprises to deploy, operate, and manage. Using a
combination of virtualization, application-level policies, and network
overlays, on-site SD-WAN devices, software platforms, or customer
premises equipment (CPE) perform two functions:
• They aggregate multiple public and private WAN links.
• They automatically select the most optimal path for traffic,
based on real-time conditions.
The latter function has historically required network managers
to manually reconfigure their networks anytime they wanted to
shape the direction of traffic over multiple routes.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 123
3.2.6. Wide Area Network and is only decrypted once it reaches its
destination. Since the WAN is establishing
(WAN)Standard
a secure channel across a public area,
An extensive geographic network, such as this is known as “tunneling.” Moreover,
the Internet, is referred to as a wide area firewalls prevent hacker intrusion. This kind
network (WAN). Smaller local area networks is probably the most widely used since it
(LANs) and metropolitan area networks offers excellent advantages at a reasonable
(MANs) are contrasted with this. A MAN price. It has a high transfer rate and is
may cover a campus or provide services to an “always on” connection, so the WAN
city residents, as in the case of a citywide is always available. The most affordable
wireless or Wi-Fi network, whereas LANs kind of WAN connects to the Internet via
are networks found in homes or offices. a dial-up modem. Since DSL has dropped in
Although the Internet is a public wide area price to the point where dial-up accounts
network (WAN), private or business models can now compete, this is less common. A
can be developed in a variety of ways. In normal DSL connection can operate at up
essence, a private network consists of two to 20 times the speed of a dial-up modem,
or more LANs linked together. which can only reach 56 kilobits per second
In order to facilitate communication (kbps). Additionally, dial-up connections are
between LANs connected to a WAN, routers not able to share phone service. Lastly, a
are used. Each data packet that travels over dial-up connection is not “always on.” The
the WAN is read by the router, which is effective reduction of uptime can occur when
installed on the leased line. It then forwards offices are located in different time zones.
the appropriate LAN to the appropriate A wide area network (WAN) is a great tool
packet’s “envelopes,” or headers. A device for businesses to centralize productivity and
known as a switch forwards the data packet access geographically dispersed resources.
to the appropriate computer when it reaches A rented space.
the local area network (LAN). Therefore,
for long-distance communication, the WAN Did you know?
functions as an interface between LANs. WANs are often built using leased lines. At each
end of the leased line, a router connects the LAN on
As there is no public traffic on a leased one side with a second router within the LAN on the
line, one that operates on it is considered other. Because leased lines can be very expensive,
private. Many businesses that need a WAN instead of using leased lines, WANs can also be
instead use an Internet Service Provider built using less costly circuit switching or packet
(ISP) to provide access because leased lines switching methods.
are costly. In this instance, a typical digital
subscriber line (DSL) account is used for
communication between each LAN in the 3.2.7. WAN Port
WAN. An existing phone line is shared The local area network (LAN) and the wide
with the DSL Internet account and used area network (WAN) are the two main zones
for phone calls. that make up most networks. A local area
network, or LAN, can be found inside any
Public Internet lines are used by a DSL- type of building, be it a two-computer
based WAN. It is secured against intrusion home or a high-rise office complex housing
by establishing a Virtual Private Network thousands of computers. The network
(VPN). When using a VPN, all communication outside of the local area network (LAN)
over the Internet is encrypted while in transit includes the entire Internet as well as other
124 COMPUTER NETWORKS

internal networks. The gateway that allows example of this would be that three other
data to move back and forth between a LAN routers are connected to a single router via
and a WAN is called a WAN port. its LAN ports, and it receives high-speed
information from those routers at its WAN
On a network router, WAN ports are ports. With respect to one another, these
typically located for users. Four LAN ports internal networks would be completely
and one WAN port are found on a typical separate entities with no more connection
home router. For the WAN port, they are than if they were part of a network in an
referred to by some routers as an uplink, entirely different building.
and for the LAN ports, as wired connections.
This port splits data to multiple computers
within the home network after receiving it 3.2.8. Build a WAN
from a high-speed Internet source, like a
Creating a Wide Area Network, or WAN,
cable modem. Although four LAN ports are
entails setting up a link to your location
found on most home routers, there can be
through a third-party service provider.
zero, an infinite number, or rarely more than
Various WAN technologies can be used
eight. All five of these ports are visible as
based on your location and needs. For
rectangular holes on the device’s back. They
instance, much like a VPN, an MPLS network
are all made to be connected to a network
can be used to establish a single network
cable. Though it otherwise has the same
from several physical locations. You need
appearance as the LAN ports, the WAN
your own networking hardware, such as
port is typically positioned differently from
switches and routers, and a service provider
the other ports to facilitate identification.
contract in order to construct a wide area
The router is where the two port types network (WAN).
differ most significantly. Information from the
1. To find out what kinds of WAN
Internet or an external network is received
services are available, get in touch
by the WAN port. The internal firewall
with a local service provider. Frame
and routing system of the router filter the
relay and T1 are common services.
data. After that, the data is transmitted to
It’s possible that business-class
the appropriate LAN port or to a wireless
cable and DSL services are available
source via a wireless connection. Routers
in your area. The service provider
have switching capabilities in addition to
establishes a demarcation point
routing and firewalling capabilities. This
and installs their equipment at
enables communication between computers
your location after you choose a
that are linked solely through LAN ports.
desired service.
The PCs are all connected to the same
network because this switching feature gets 2. Get a router and attach the WAN
around the router’s built-in firewall. Users connection to it. Even if the service
can expand the size of their network by provider installs a router as part
using this feature to link multiple routers of the wide area network (WAN),
to one another. you will probably still require
a separate router for local area
Multiple internal networks that function network (LAN) access and control.
independently of one another could be It is your responsibility to use a
created if they connected several routers router that can connect to the WAN
to one another via the WAN port. An circuit if the service provider does
 LOCAL AREA NETWORK/WIDE AREA NETWORK 125
not supply one. For instance, your router needs to have a T1 interface module
if you bought a T1 connection.
3. Link your router to the network switch. All LAN connections are gathered by
a network switch and connected to the router, which then routes the relevant
traffic over the WAN connection. Using one or more Ethernet cables is the most
typical method of connecting a switch to a router. Your router might have a
built-in switch with enough ports to connect every device on it, depending on
how your network is set up.

Did you get it?

1. What functions and features do WAN devices typically possess?
2. How are WAN ports configured and utilized in wide area network setups?
126 COMPUTER NETWORKS

A CLOSER LOOK
Local Area Networks (LANs) and Wide Area Networks (WANs) are two fundamental types
of computer networks, each serving distinct purposes and catering to different scales
of connectivity. Despite their shared goal of facilitating communication and resource
sharing, they exhibit several significant differences upon closer examination.
LANs are typically confined to a limited geographical area, such as a single building,
office, or campus. They are characterized by high data transfer rates and low latency,
making them ideal for facilitating fast communication and resource sharing among devices
within their defined boundaries. LANs commonly utilize Ethernet or Wi-Fi technologies
for connectivity, with devices interconnected through switches or wireless access points.
Due to their localized nature, LANs are relatively easier to set up, manage, and maintain
compared to WANs. They offer inherent security advantages as the network administrator
has more control over access and monitoring within the confined environment.
In contrast, WANs span larger geographical areas, often connecting multiple LANs
or individual devices across vast distances, such as between different cities, countries,
or continents. WANs employ various technologies, including leased lines, satellite links,
and fiber optics, to establish connections over long distances. The expansive reach
of WANs introduces challenges such as higher latency and lower data transfer rates
compared to LANs due to the increased distance and the involvement of multiple
interconnected networks and devices. WANs are commonly used by organizations to
facilitate communication between geographically dispersed branches, enable remote
access for employees, and access centralized resources hosted in data centers.
One of the key differences between LANs and WANs lies in their ownership and
management structures. LANs are typically owned, operated, and maintained by a
single organization, such as a company, university, or government agency, allowing for
centralized control and customization according to specific requirements. In contrast,
WANs often involve collaboration between multiple entities, including Internet Service
Providers (ISPs), telecommunications companies, and government agencies, making
their management more complex and requiring adherence to standardized protocols and
agreements for interconnection and data exchange.
Another critical aspect to consider is scalability. LANs are inherently scalable
within their confined environment, allowing for the addition of devices and expansion
of network capabilities as needed. However, scaling a LAN to cover larger geographical
areas would require transitioning to WAN technologies and addressing the associated
challenges of long-distance communication and connectivity. WANs, on the other hand,
are designed to scale across vast distances, accommodating the growth and changing
needs of organizations operating across multiple locations or serving diverse user bases.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 127

CASE STUDY
CASE STUDY AND ANALYSIS OF WAN OPTIMIZATION
PRE-REQUIREMENT
In today’s world, the WAN network is congested with high traffic, which causes a lot
of data loss and jitter. In order to provide a solution for this, a WAN Optimizer (e.g.,
Riverbed) was introduced over the WAN network to significantly improve network
performance. However, there are many criteria that must be fulfilled to achieve a
considerable amount of data transaction, including a decrease in the percentage of data
loss over the network. These criteria will be analyzed during the discussion of the case
study. Every organization has multiple applications running over the network, resulting in
the available bandwidth being shared among them. If not balanced properly, this could
lead to significant losses in business, data transactions, and application performance.
Therefore, it is important to design this network in a way that ensures the organization’s
investment is satisfactory.

I. Criteria’s involved in WAN Optimization

Whenever the concept of WAN is taken there are pre requisites that are taken into
account like Network Infrastructure, Network Design (Head Office with other Branches),
Cable used (Cat 5e, 6 within Organization, Fiber Optic Cable connecting branches, core
switches), Model of Router, Model of Switches (Layer 2 or 3), Capacity of the organization,
Application that will work on live environment which includes network security like
ISA, Forefront, Business solution, etc., Servers, Data Base. How the calculations among
them are carried out?
i) Network Bandwidth Consumed = TB – TC (1)
TB – Total Bandwidth provided by ISP
TC – Total bandwidth utilized by the organization during peak time and weekends/
non-working hours
ii) % of data loss = Data Loss / Total data transmitted *100 (2)
E.g. Data Loss = 5000Kbps Data Transmitted = 2500 Kbps
5000/2500 * 100 = 200% (Ratio becomes 2:1)
iii) Ratio between % of bandwidth consumed in internal network: % of bandwidth
consumed over external network (This provides us the idea on how to manage
the network performance during peak hours based upon which the prioritization
should be done) (3)
iv) Network Performance
= Data transmission over Fiber optic + Data Transmission over Cat 6/5e
Cables (4)
128 COMPUTER NETWORKS

Data Transmission = Total data transmitted – Data loss (Calculate this for Fiber
Optic and Cat 6/5e cables) (4(i))
% of network performance = Total Data transmission/ Total bandwidth provided
* 100
v) % of service provided by ISP = (24*7 – Downtime) / (24*7) * 100 (5)
% of Service provided by ISP: % of overall Output (Expected along with fault
tolerance)
vi) Total Consumption (%) = ABC*UBC*BW*DBC*UWBC (6)
ABC – Application Bandwidth Consumption UBCUser Bandwidth consumption
BW- Bandwidth wastage DBC- Data Transaction Bandwidth Consumption UWBC
– Unwanted Bandwidth Consumption.
vii) % of N/W used during peak hours : % of N/W usage in non-peak hours/ weekend
These are some of the criteria that will help the Infrastructure team design the
network initially and put some of the operations of an application live to get an idea
of the network consumption as a whole. Based on the overall performance, they can
configure the WAN Optimizer, which will in turn accelerate the performance of the
applications running live.

Figure 1. WAN optimization and performance (Source: Bhargav. Balakrishnan; Case study and
analysis of WAN Optimization pre-requirements; https://arxiv.org/ftp/arxiv/papers/1109/1109.2285.
pdf).

For example, Finacle is now widely used in the banking sector, where it will be
integrated with Oracle Applications, Oracle Database, etc. Along with this, there are
commonly used applications like Active Directory service, Exchange Servers, Anti-
 LOCAL AREA NETWORK/WIDE AREA NETWORK 129

Virus Servers, File Servers, SharePoint Servers, Web Servers, Blackberry Servers, etc.
Therefore, the bandwidth consumed by these applications should be carefully studied
by the network management, server management, and application management teams.
Based on this, the router and WAN Optimizer should be configured to achieve the best
overall output by minimizing data loss. Once data loss is reduced, transmission over
the network becomes much faster, thereby reducing jitter and data loss to a greater
extent. However, it is not possible to eliminate data loss and jitter completely, as there
is always a certain amount of data loss and jitter, along with a reduction in network
performance.
II. Case Study of Bank (Large Sectors)

Let us consider a Bank with 200 employees and three branches:

1. Head Office (Muscat) – 75
2. Sur Branch – 25
3. Sohar – 60
4. Nizwa – 40
So the above-mentioned applications in the topic criteria involved in WAN Optimization
will be working along with certain other Backup Solution and Data Archiving, i.e., Veritas
NetBackup and NetVault. Here, the Head Office will mostly have all the servers running
the applications mentioned above. So, how is the network going to be maintained
optimized, thereby providing the best results at the end of the financial year? The
network diagram for this is shown below.
According to the network diagram above, the branches will depend on the services
from the head office. Therefore, the prioritization of applications will come into play. In
countries like Oman, the telephone exchange outside Muscat mostly provides a maximum
bandwidth of 128Kbps leased line because the exchange can only support that amount of
bandwidth. In this case, prioritization will have to be given to Finacle and Mail Servers
as they are the main applications needed for processing data transmission with the
clients, ensuring that reports can be generated without any issues by the end of the
day. Internet access permissions can be granted for managerial level employees as they
require certain other applications to communicate with the head office, such as Office
Communicator and instant messenger. This way, application usage is balanced so that
all data storages, transactions, and report updates are generated properly for the entire
organization. This demonstrates how well the network is optimized by using a WAN
Optimizer, which ultimately leads to organizational satisfaction with their investment,
commonly referred to as return on investment.
To achieve this, the IT team should design the network in such a way that the
above result is obtained constantly. The network design should incorporate redundancy,
such as an MPLS connection that provides 24-hour support from the service provider,
along with fault tolerance. In case one connection fails, the other connection will take
over. However, the investment required is high for medium and small-level sectors.
130 COMPUTER NETWORKS

This consideration is essential to ensure uninterrupted service in case of unplanned

downtime by the ISP. The combination of the above information provides a solution on
how to structure a network, prioritize, and implement the usage of WAN Optimizer. The
entire IT team and managers should create a workflow chart on how to achieve the best
throughput by optimizing the speed of applications. This falls under service transition,
and any changes in applications, such as switching from Finacle to Fin Flex if it is
lighter and more stable, should be considered based on the bank’s requirements. After
finalizing the report, the IT team should proceed with designing, testing (application
in live environment), and execution.
A. Designing Phase & Test

During the design phase, the IT Team will have all the reports that were discussed.
Initially, the team should start working on the core applications that will be widely used
by the organization. Subsequently, all other common applications should be taken into
account, and a statistical graph should be created to show the approximate bandwidth
consumption by each application. An example of a statistical graph is shown below. In
this graph, let’s consider the consumption of Sharepoint portal and Finacle, which are
used more frequently. Therefore, these applications should be prioritized based on their
importance to the organization. These applications will be configured on WAN optimizer
to ensure performance can be maintained. Backup should also be taken care of in WAN
Optimization as it plays a role in storing and retrieving data. Therefore, these specific
applications should be tested before being launched live. The calculations of the above
formulas should be done in both test and live environments so that the IT team can
have a clear understanding of the bandwidth consumption. These calculations can serve
as a case study for future upgrades and references. Based on this, the IT Team will be
able to determine how successfully they have implemented the infrastructure for the
organization.
B. Live Environment

The organization should always conduct periodic tests on the performance of all applications
over the network. Any identified flaws should be promptly analyzed and the information
should be sent to the respective team. Based on this, the change advisory board should
decide on an alternate solution, for example, if the bandwidth consumption is high during
peak hours. The reasons for these issues should be discussed during meetings, such as:
1. Users downloading unwanted files from the internet 2. Certain applications utilizing
excessive bandwidth, especially core applications 3. Data storage transmission between
applications and the database 4. Anti-virus and firewall applications like ISA, Forefront
Security, etc. These are common issues that need proper justification. To justify these
issues, the team should prepare a case study on the application and conduct testing from
all perspectives. For example, if it involves a database, consider how data is distributed
over the network, such as network load balancing, to avoid performance issues. When
structured in this manner, WAN Optimizer can enhance application performance. To
achieve the best throughput with WAN Optimizer, certain prerequisites need to be met.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 131

It is crucial to balance the performance of all applications over the network, especially
for databases that primarily work on clustering.
The test results of all live servers should be presented to management every six
months as a report to demonstrate how well the infrastructure is maintained within
the organization.
C. Execution

This will be final phases after Live Testing of the critical applications. Here the reports
of all the testing which were done in the live servers will be verified that the condition
satisfies the usage of WAN optimizer and the respective applications for which this
needs to be prioritized. How is the prioritization calculated?
1. Prioritization in % = Number of usage * Approx. time duration utilized /total
no of employee * 100
In approximate time duration time duration of peak hours with + or – extra
hours.
2. Downtime in % = Total working hours(Application in WAN optimizer) / Total
working hours *100 Downtime is unplanned downtime. Maintenance will be
reduced in total working hours itself.
3. Percentage of transactions = Total of transaction done – Total no of lost transaction
/ Actual number of transaction * 100
So why above calculation are done for fixing a WAN Optimizer? What is need of
these testing processes? Hardly the device is going to cost some 10 Lakhs Maximum
with installation charges. The answer for this every inclusion of additional device needs
a proper justification and at the time same the investment should have some output
in the organizations overall result. These tests are conducted solely to ensure that the
critical applications are brought under the umbrella of the WAN Optimizer, which will
handle the performance of all applications configured under it. Subsequently, these
tests should be conducted to verify if there are any changes in the processing speed of
the desired applications. This will generate the final report for the management team,
indicating that WAN optimization will enhance the performance of critical applications
within the organization. Consequently, the organization can anticipate these results in
the upcoming financial year.
When the Excel sheet containing the above graph is opened, there is a prioritization
of core applications used in the banking sector. This scenario will be the same for other
sectors as well. Therefore, when the WAN Optimizer is configured in this manner, the
performance of the core applications will significantly improve, especially the daily backup
of the entire banking process. It should be generated properly to avoid data loss in case
of network performance interruptions, as the data transfer percentage depends on the
type of cable used, particularly for backup purposes where fiber optic cables are utilized.
In countries like India, the end-of-day (EOD) process can be cumbersome, making this
optimization process quite relieving, as it reduces the need for constant monitoring.
132 COMPUTER NETWORKS

Figure 2. Live testing of WAN (Source: Bhargav. Balakrishnan; Case study and analysis of WAN
Optimization pre-requirements; https://arxiv.org/ftp/arxiv/papers/1109/1109.2285.pdf).

III. Present Scenario

Many organizations have encountered WAN Optimization. However, to what extent

have they been analyzed and utilized for organizational purposes? How satisfied are
end users with the performance? Even though the network is optimized to enhance
application performance, it depends on whether they use a wired or wireless network.
This also significantly impacts application performance. Additionally, branch offices
located in remote areas must also be taken into consideration. All these aspects need
to be analyzed before conducting live testing. Once these network setups are confirmed,
implementation becomes the next step. In the current scenario, how extensively are
these reports prepared by the IT team to support the adoption of organizational policies?
To what extent do organizations adhere to IT policies? Various policies such as group
policies, internet restrictions, and USB restrictions need to be established to fully leverage
WAN optimization benefits.
These policies need to be established, and comprehensive network testing must be
conducted. Based on the results, a report should be generated. This report will then
be reviewed by the change advisory board and approved, indicating that there are 99%
fewer flaws and no fault tolerance in the current network infrastructure. Achieving this
level of accuracy is crucial. However, many large organizations struggle to guarantee it.
Achieving this is a complex task that requires time for testing to ensure that the network
is optimized and that application performance is effectively shared across the network,
reducing data loss and jitter compared to previous records without WAN optimization.
IV. Process Flow Diagram

The process flow of this implementation should be according to the above Step that,
i.e., designing and test.
 LOCAL AREA NETWORK/WIDE AREA NETWORK 133

This workflow will give an idea of HOW ORGANIZED AN IMPLEMENTATION IS

TAKEN. Since the implementation of WAN Optimization is interrelated with other
networking devices, the notification should be generated as an email or SMS (SMS
through Blackberry Enterprise Server). The clarification needs to be clearly documented
for future reference if the organization is not benefiting in any way from the WAN
Optimizer. This workflow involves proper approval at each stage by the IT Manager or
CAB Member (Involves SLA Manager, Clients, Business Relationship Manager). So this
kind of workflow will help in executing a task effectively and with proper authentication.

Figure 3. Process Flow Diagram of WAN (Source: Bhargav. Balakrishnan; Case study and analysis
of WAN Optimization pre-requirements; https://arxiv.org/ftp/arxiv/papers/1109/1109.2285.pdf).
134 COMPUTER NETWORKS

V. Key Advantage on Various WAN Optimizer In today’s world there are various
WAN optimizer and each of them have their own features let us some of them like
Riverbed, Bluecoat, Juniper and Cisco.
A. Bluecoat

Works on the protocol CIFS (Common Internet File System) which can significantly reduce
the number of round-trips required to satisfy a request, effectively minimizing delays
associated with waiting for data retrieval while simultaneously reducing WAN usage.
Another major advantage of CIFS protocol is CIFS Protocol Optimization is possible
because ProxySG appliances have the ability to terminate user requests as if they were
the server. They can then open a separate connection to the server that they can use to
intelligently make requests on the client’s behalf. The ProxySG appliance, acting as the
client, is able to take advantage of the CIFS protocol in a way that the client cannot.
B. Riverbed

Riverbed’s award-winning wide-area data services (WDS) solutions strategically enable

IT to centralize and reduce operational overhead and expenses, while improving end-
user satisfaction. Riverbed’s products have been proven in some of the most demanding
and complex networks in the world, with thousands of customers deploying Steelhead
WDS solutions. Some key advantages of Riverbed Steelhead are speed, scalability, and
accelerated key enterprise applications. Above all, there are some major advantages:
IT infrastructure consolidation, serverless branch office, enhanced data protection, and
enabling the mobile workforce. Key streaming processing data, application, transport,
and management will improve the performance of key enterprise applications by 30%-
40% compared to before. For example, if the current performance is 60%, it will be
improved to 95% by using the Riverbed Steelhead.
C. Cisco

Cisco’s WAAS has introduced the WAN Optimizer known as Wide Area Application
Services. This offers significant advantages in terms of deployment and management
compared to other WAN optimizers. The WAAS Central Manager is a web-based
central management tool that simplifies configuration, provisioning, monitoring, fault
management, logging, and reporting for up to 2,500 WAEs within a Cisco WAAS topology.
Cisco’s WAAS does not require any modifications to applications, clients, or servers to
provide acceleration services.
Transparency: Transparency is crucial for many IT organizations when implementing
a WAN optimization solution. It is essential that the network’s functionality, such as
security, QoS, or routing, remains intact. With WAAS not altering packet headers, it
ensures a high level of transparency. This transparency impacts the Total Cost of
Ownership (TCO) significantly.
Ease of Deployment and Management: Ease of Deployment and Management is another
key benefit of Cisco’s WAAS. By not modifying packet headers, WAAS minimizes the
 LOCAL AREA NETWORK/WIDE AREA NETWORK 135

risk of disrupting IT operations, making it easier to deploy and manage compared to

other solutions that may cause disruptions.
Moreover, WAAS supports auto-discovery, which checks for a peer acceleration appliance in
the packet flow between the source and destination. This feature streamlines deployment
by eliminating the need for IT organizations to set up an overlay network.
Integration: The point being that given Cisco’s position in the enterprise networking
marketplace, the chances are that the environment in which the WAN optimization
solutions will be deployed will be a Cisco environment. Cisco is in the best position to
ensure that its WAN optimization solution does not impact this environment.
Performance: Some of the comments that Miercom made about Cisco’s WAAS include:
o Under favorable conditions (large WAN latency and highly compressible content)
the WAAS V4.0 solution can provide impressive degrees of acceleration.
o The software demonstrated performance parity with these other products, and
that in some key metrics provided superior compression, speed, and throughput.
o Unlike some of the other products in the market, Cisco’s WAAS does not degrade.
So each WAN Optimizer manufacturing company tests their device on the aspect of
fault tolerance, i.e., preventing any disturbance in the current infrastructure setup.
In the same way, this case study needs to be carried out by the IT team in order to
justify to management why they have selected this WAN optimizer and what the key
advantage of that is.
136 COMPUTER NETWORKS

SUMMARY
• A local area network (LAN) is a computer network in a limited physical area
like a house, school, office building, or computer lab.
• A local area network, or LAN, is a type of network that can operate over a
small physical area, like a few kilometers in size, which could be an office,
factory, or collection of buildings.
• Local area networks (LANs) are widely used in the information age due to
the advent of low-cost computers and the need for hardware resources and
information sharing in office settings, leading to the development of affordable
LANs.
• A server on a network provides services to workstations and clients. The main
goal of a LAN server is data management. It ensures data is stored, secured,
and retrieved.
• Servers are business-class PCs that offer collaboration tools, antivirus protection,
messaging, and other services to the office.
• A local area network, or LAN, is a type of data communication network that
links computers or terminals within a building or a constrained geographic area.
• Frame Relay is a WAN protocol that is only for internetworking Local Area
Networks (LANs), or the transfer of data between LANs and WAN endpoints.
It is based on packet switching technology.
• The connection between your router and the service provider must be made
before you can use any type of WAN connection. The type of connection will
determine how you proceed.
• A wide area network (WAN) is a network that connects smaller campus and
metropolitan networks across different locations into a single, distributed network
using various links such as private lines, virtual private networks (VPNs),
wireless (cellular), MPLS, and the Internet.

MULTIPLE CHOICE QUESTIONS

1. Computer network that spans a limited physical area, usually ranging from a small
office to a building is known as
a. LAN
b. WAN
c. MAN
d. PAN
2. Computer network which spans a large physical area, connecting several sites of
an organization across cities, countries and continents is known as
a. LAN
b. MAN
 LOCAL AREA NETWORK/WIDE AREA NETWORK 137
c. WAN
d. PAN
3. Most common type of a computer network is
a. PAN
b. WAN
c. LAN
d. MAN
4. Modulator and demodulator as combine is known as –
a. Modulus
b. Modem
c. Mod switch
d. Mod access
5. Which of the following is an example of Personal Area Networking?
a. Bluetooth
b. WAN
c. WLAN
d. All of the above

REVIEW QUESTIONS
1. What are the characteristics of LAN technology?
2. What hardware devices are typically found on a LAN?
3. What are WAN devices and how do they differ from LAN devices?
4. How do you connect your Wide Area Network (WAN) devices?
5. How can WAN management and optimization be achieved?

Answer to Multiple Choice Questions

1. (a) 2. (c) 3. (c) 4. (b) 5. (a)

REFERENCES
1. Campbell, S., (2010). A Wide Area Network (WAN) Tutorial [Technology Explained].
[online] MakeUseOf. Available at: http://www.makeuseof.com/tag/technology-
explained-wide-area-network-wan-tutorial/ (accessed on 05 April 2024).
2. Computerhope.com. (2016). What is WAN (Wide Area Network)? [online] Available
at: http://www.computerhope.com/jargon/w/wan.htm (accessed on 05 April 2024).
138 COMPUTER NETWORKS

3. Mitchell, B., (2015). How ‘Packet Switching’ Works (Computer Networks).

[online] About.com Tech. Available at: http://compnetworking.about.com/od/
networkprotocols/f/packet-switch.htm (accessed on 05 April 2024).
4. Mitchell, B., (2016). What Is a Wide Area Network (WAN) and How Does it Work?
[online] About.com Tech. Available at: http://compnetworking.about.com/cs/
lanvlanwan/g/bldef_wan.htm (accessed on 05 April 2024).
5. Point to Point WAN: The Most Effective Way for Branch to Branch Transaction
and Communication, (2011). [Blog] Point to Point WAN. Available at: http://
pointtopointwan.blogspot.com.au/ (accessed on 05 April 2024).
6. Symoens, J., (2000). Preparing Exchange for High Availability-You Can Increase
Uptime and Decrease Costs by Building a Redundant Groupware Configuration.
InfoWorld.
7. Winther, M., (2006). Avoiding the challenges of do-it-yourself broadband VPNs.
Business Communications Review.
CHAPTER

4 Internetworks

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Explain the principles of internetworking

• Define the internetwork architecture
• Discuss on IP addressing and architecture
140 COMPUTER NETWORKS

INTRODUCTION
The original purpose of internetworking was to link different forms of computer networking
technology. The term “computer network” refers to a group of two or more connected
computers. Computer internetworking involves connecting two or more computer LANs,
WANs, or network segments using routers and configuring them with logical addressing
schemes and protocols like IP. The Internet Protocol is commonly used by interconnected
computer networks, also known as Internetworking. There are two architectural models
frequently used to explain the procedures and protocols in internetworking. Open Systems
Interconnection is the industry standard reference model for internetworking (OSI).
Companies like Cisco, BBN, and other network product and service providers use the
term “internetworking” to encompass all the concepts, technologies, and everyday items
that allow people and their computers to communicate over different types of networks.
For example, a user on a computer connected to a token ring local area network might
want to communicate with a user on a computer connected to an Ethernet local area
network in another country using a wide area network interconnection. Internetworking
consists of standard internetwork protocols, relevant network devices, and routing tables
necessary for this communication to take place.
Open Systems Interconnection (OSI) is the standard reference model for internetworking
and can also be used as a model for intranetworking. Because all technologies can be
related to the standard communication model, OSI makes it possible for any technology
to be related to another technology. In order to make the problem of exchanging data
across a network or a network of networks easier to understand and to make defining the
boundaries between components easier, OSI offers a layering approach to the problem.
4.1. PRINCIPLES OF
INTERNETWORKING
→Learning Objectives
• Understand the fundamental principles underlying internetworking
• Recognize the various challenges encountered in building and managing
interconnected networks.
• Define internetwork addressing and its significance in facilitating communication
between devices across interconnected networks.
• Explain the concept of a unit of internetworking and its role in establishing
connections and data exchange between devices.

The term “internetworking” is a compound of the words “inter” and “networking,”

which denotes a connection between completely unrelated nodes or segments. The
intermediary devices, which are similar to routers or gateways, establish this connection
area unit. Catenet was the initial term for an internetwork. This connection frequently
occurs within or between networks that are governmental, commercial, industrial, public,
or private. As a result, an internetwork may consist of several small networks connected
by intermediary networking devices that work together to form a single, large network.
The term “internetworking” describes the services, goods, and methods that address
the difficulty of creating and managing internetworks.
Every network node and phase is built with a similar protocol or communication
logic, such as Internet Protocol (IP) or Transmission Control Protocol (TCP), to enable
communication. Internetworking is the process of a network communicating with another
network through continuous communication protocols. Internetworking was created to
address the issue of sending an information packet over numerous links. There is a subtle
distinction between Internetworking and network extension. A simple LAN extension
is when two local area networks are connected via a switch or hub; connecting them
via a router is an example of an Internetworking setup. The OSI-ISO model’s Network
Layer, or Layer Three, enforces internetworking. The Internet is the most prominent
illustration of internetworking.

4.1.1. Principle
The process of connecting multiple networks via devices like routers and gateways
is known as internetworking. These networks, owned by various organizations, use
different technologies, ranging from small LANs to massive WANs. These networks
can be public, private, government, commercial, or academic. The term “internetwork”
or “internet” is commonly used to describe this network of networks. Networking is

INTERNETWORKS 141
142 COMPUTER NETWORKS

essential for computers of all sizes to communicate with each

other, regardless of vendor, operating system, hardware platform,
or geographic location. Internetworking involves tying together
networks with different technologies using additional hardware
and software. DARPA has funded research projects to develop
standards for internetworking. Layering and other protocol suite
concepts were first introduced to the networking community by
the DARPA project ARPAnet, long before ISO took the lead in this
Keyword area. DARPA persisted in investigating a suite of internetworking
Network Control protocols. This can be observed in the early host-to-host protocol
Program (NCP) of the Network Control Program (NCP) to the TCP/IP protocol
was the name
for the software suite, which assumed its present form circa 1978. In 1975, the
on hosts which Defense Communications Agency (DCA) was given administrative
implemented the responsibility for ARPAnet, which was declared an operational
Network Control
Protocol of the network. DARPA was widely recognized for having pioneered packet
ARPANET. switching over radio networks and satellite channels. There was
no TCP/IP developed yet.
The ARPAnet was essentially a leased-line network connected by
Internet Message Processors (IMP), which are specialized switching
nodes. By 1979, TCP/IP research had attracted a large number
of researchers. This prompted DARPA to establish an unofficial
committee to oversee and direct the development of the architecture
and communication protocols.
The Internet Control and Configuration Board (ICCB) was the
name of the committee. When DARPA began upgrading the machines
in its research network ARPAnet to use the new TCP/IP protocols,
it is possible to trace the earliest actual implementations of the
Internet to that period. Following this shift, which took place
between 1980 and 1983, DARPA mandated that TCP/IP be used by
any computer that wished to connect to its ARPAnet. In 1983, the
US military made TCP/IP the standard protocol and demanded that
all networks connected to the ARPAnet follow the new guidelines.
The ARPAnet exceeded even its own founders’ expectations
in terms of success, and TCP/IP internetworking spread quickly.
Consequently, new wide area networks (WAN) were established in
the USA and linked via TCP/IP protocol to the ARPAnet. The set of
interconnected networks was then expanded to include additional
networks from around the globe, some of which were not even
based on the TCP/IP protocols.
The world’s largest network is currently made up of computing
facilities connected to the Internet via separate sub-networks from
regions such as Europe, Japan, and the United States of America.
The Internet was established as the official global network in
 INTERNETWORKS 143
1990, replacing ARPAnet. The development of a TCP/IP protocol
implementation for Berkeley UNIX on the VAX, as well as the
free distribution of the required code produced as a result of this
project along with their UNIX operating system, was also funded
by DARPA. In 1983, the TCP/IP protocol set was added to the
Berkeley Software Distribution (BSD) for the first time (4.2BSD). TCP/
IP quickly spread throughout academic institutions and research
facilities as a result, and it is now the required communications
subsystem for all UNIX connectivity. There are numerous BSD code
updates available. These include 4.3BSD (1986), 4.4BSD (1993),
4.3BSD Tahoe (1988), and 4.3BSD Reno (1990).

4.1.2. Challenges in Internetworking

There is no guarantee that a useful internetwork will be
implemented. Every area is crucial to building an affordable and Keyword
efficient internetwork, and there are many demanding fields,
Troubleshooting
especially in the areas of dependability, connectivity, network is a form of problem
management, and adaptability. A select few are: solving, often
applied to repair
• The first difficulty arises when we attempt to link multiple failed products
or processes on
systems to facilitate communication across various a machine or a
technologies. For instance, completely different websites system.
may use entirely different media or run at varying speeds.
• In an internetwork, maintaining dependable service is
another crucial consideration. Access to network resources
needs to be consistent and dependable for both individual
users and entire organizations.
• In an internetwork, network management should provide
centralized support and advanced troubleshooting
capabilities. The internetwork must have its configuration,
security, performance, and other issues properly addressed
for it to function properly.
• The most crucial aspect, flexibility, is crucial for a variety
of reasons, including network expansion and the addition
of new services and applications.

4.1.3. Internetwork Addressing

Devices can be established individually or as a group using
internetwork addresses. Depending on the OSI layer and the protocol
family, different addressing schemes apply. Media Access Control
(MAC) addresses, network-layer addresses, and data-link layer
addresses are the three types of internetwork addresses that are
typically used.
144 COMPUTER NETWORKS

4.1.3.1. Data Link Layer Addresses

Every physical network association of a network device is uniquely identified by its
data-link layer address. Physical or hardware addresses are commonly used to refer to
data-link addresses. Data-link addresses are associated with specific devices and are
sometimes found within a flat address area. Their relationship is pre-established and
typically fastened. End systems typically have a single data-link address because they
have a single physical network association. Due to their multiple physical network
connections, routers and other internet-connected devices typically have multiple data-
link addresses.
Every physical network connection made by a network device is uniquely identified
by its data-link layer address. Sometimes, physical or hardware addresses are used to
refer to data-link addresses. A pre-established and usually fixed relationship between
a data-link address and a particular device exists within a flat address space. End
systems typically have one data-link address and one physical network connection. Since
routers and other internet-connected devices frequently have several physical network
connections, they also frequently have several data-link addresses. A data-link address
serves as a unique identifier for each interface on a device, as shown in Figure 4.1.

Figure 4.1. Each interface on a device is uniquely identified by a data-link address (Source:
Scribd, https://image.slidesharecdn.com/internetworking-49-121020113333-phpapp02/75/
internetworking49-15-2048.jpg?cb=1667679119).

4.1.3.2. MAC Addresses

A collection of data-link layer addresses is included in media access management
(MAC) addresses. In LANs that use the IEEE MAC addresses of the data-link layer,
MAC addresses define network entities. For each local area network interface, MAC
addresses vary in specific ways. Twelve hexadecimal digits are used to express a MAC
 INTERNETWORKS 145
address, which has a length of 48 bits. The Organizational Unique Identifier is made
up of the first six hexadecimal digits, which are typically used by the IEEE to identify
the manufacturer or merchant (OUI). The final six positional notation digits represent
either the interface serial variety or an additional price managed by the specific vendor.
Because they are burned into read-only memory (ROM), MAC addresses are also known
as burned-in addresses (BIAs). After the interface card initializes, MAC addresses are
traced into random-access memory (RAM).
A subset of data-link layer addresses makes up media access control (MAC) addresses.
In LANs that use the IEEE MAC addresses of the data-link layer, MAC addresses are
used to identify network entities. Every LAN interface has a unique MAC address, just
like the majority of data-link addresses. The relationship between data-link addresses,
MAC addresses, and the IEEE sub-layers of the data-link layer is depicted in Figure 4.2.

Figure 4.2. MAC addresses, data-link addresses, and the IEEE sub-layers of the data-link layer
are all related (Source: Scribd, https://image.slidesharecdn.com/internetworking-49-121020113333-
phpapp02/75/internetworking49-19-2048.jpg?cb=1667679119).

MAC addresses are expressed as 12 hexadecimal digits and have a length of 48 bits.
The manufacturer or vendor is identified by the first six hexadecimal digits, which are
managed by the IEEE, and together they make up the Organizational Unique Identifier
(OUI). The interface serial number, or any other value managed by the particular vendor,
is contained in the final six hexadecimal digits. Because MAC addresses are burned into
read-only memory (ROM) and copied into random-access memory (RAM) upon interface
card initialization, they are sometimes referred to as burned-in addresses (BIAs). The
format of the MAC address is shown in Figure 4.3.
The techniques used by various protocol suites to ascertain a device’s MAC address
vary. MAC addresses are mapped to network addresses via Address Resolution Protocol
(ARP), which is the most widely used of the three techniques. Network devices can
discover the MAC addresses of other network devices thanks to the Hello protocol. MAC
addresses can be produced by an algorithm or they can be included in the network-
layer address.
Mapping network addresses to Media Access Control (MAC) addresses is known as
address resolution. Several protocol suites implement the Address Resolution Protocol
(ARP), which is used to carry out this process. The network device saves the data
in the ARP cache after a network address and MAC address are successfully linked.
146 COMPUTER NETWORKS

Because the destination’s MAC address is that are connected by a router, address
already known, devices can send traffic resolution functions differently. To find out
there without generating ARP traffic thanks End System Z’s MAC address, End System
to the ARP cache. Y sends out an ARP request across the
LAN. Every device on the LAN receives and
processes the broadcast, including Router
X, which serves as a proxy for End System
Z by determining that End System Z is
situated on a separate LAN by consulting its
routing table. After that, Router X responds
to End System Y’s ARP request by sending
an ARP reply that appears to be from
End System Z and includes its own MAC
address. After receiving the ARP reply, End
System Y stores Router X’s MAC address
in its ARP cache and adds it to the entry
for End System Z. In order to send a frame
directly to End System Z without utilizing
ARP requests, End System Y first looks up
Router X’s MAC address in the ARP cache.
Figure 4.3. The MAC address contains a unique
Traffic from End System Y is received by
format of hexadecimal digits (Source: Wikipedia,
https://upload.wikimedia.org/wikipedia/
Router X, which then forwards it to End
commons/thumb/9/94/MAC-48_Address. System Z on the opposite LAN.
svg/800px-MAC-48_Address.svg.png).
A network-layer protocol called Hello
Depending on the network environment, allows devices connected to a network to
there are minor differences in the address recognize one another and signal that they
resolution process. When End System A are still operational. For example, a new end
broadcasts an ARP request onto the LAN system broadcasts Hello messages across the
in an effort to find out the MAC address network upon powering up. Subsequently,
of End System B, address resolution on a devices connected to the network respond
single LAN starts. All of the LAN’s devices with “Hello,” and to indicate that they are
receive and process the broadcast, but still operational, Hello messages are also
only End System B responds to the ARP sent at set intervals. Network devices can
request by giving End System A its MAC examine Hello-protocol packets to discover
address in an ARP reply. After receiving the other devices’ MAC addresses.
response, End System A stores End System Three MAC addresses are known
B’s MAC address in its ARP cache. Network to function. Because the network layer
addresses and MAC addresses are linked either includes the MAC address in the
in the ARP cache. When End System A network-layer address or determines
needs to communicate with End System B, it the MAC address through an algorithm,
retrieves System B’s MAC address from the MAC addresses in these protocol suites
ARP cache, sends the frame straight, and are predictable. Xerox Network Systems
avoids using an ARP request in the process. (XNS), Novell Internetwork Packet Exchange
However, when source and destination (IPX), and DECnet Phase IV are the three
devices are connected to separate LANs protocols.
 INTERNETWORKS 147
4.1.3.3. Network-Layer Addresses
Network addresses, also known as virtual or logical addresses, can occasionally be found
within a gradable address space. A network address and a device have a logical, non-
fixed relationship that typically depends on either the physical properties of the network
or on arbitrary groupings. For each network-layer protocol that end systems support,
one network-layer address is required. For each supported network-layer protocol,
routers and other internetworking devices require a single network-layer address for
each physical network association.
An entity at the OSI layers’ network layer is identified by its network-layer address.
Network addresses, also known as virtual or logical addresses, are typically found in a
hierarchical address space. A network address and a device have a logical, non-fixed
relationship that is usually based on either physical network characteristics (e.g., the
device is on a specific network segment) or groupings for which there is no physical
basis (e.g., the device is a part of an AppleTalk zone). For each network-layer protocol
that end systems support, one network-layer address is needed. This presupposes that
there is just one physical network connection on the device. For each network-layer
protocol supported by routers and other internetworking devices, one network-layer
address is needed for each physical network connection. For instance, a router with
three interfaces that are each running OSI, TCP/IP, and AppleTalk needs three network-
layer addresses for every interface. Nine network-layer addresses are thus available to
the router. The process of assigning a network address to each network interface for
every supported protocol is shown in Figure 4.4.

Figure 4.4. Each network interface must be assigned a network address for each protocol
supported (Source: NotesforMsc, https://notesformsc.org/wp-content/uploads/2018/04/Network-
Addresses.png).

4.1.4. Unit of Internetworking

The idea behind internetworking is to “build a large, global network by interconnecting
different types of networks” so that any two hosts that are connected can exchange
packets. A standardized protocol defining the format and handling of transmitted
packets; components connecting the participating networks by routing packets to their
destinations based on standardized addresses; and a standardized scheme to address
packets to any host on any participating network are required to build an internetwork.
148 COMPUTER NETWORKS

There are chiefly three units of Internetworking:

• Extranet;
• Intranet; and
• Internet.
The internet is not considered to be a part of the computer
network or extranet, although it can serve as a portal for accessing
Remember parts of an extranet.
Intranets and
extranets might • Extranet: An extranet is a network within a network that
or might not is limited to a single organization or entity, but it also has
have connections restricted connections to networks of one or more other
to the net.
If there is a
organizations occasionally, but not always. It is the most
connection to the basic form of internet work, typically confined to very
net, the computer private spaces. An extranet cannot consist of a single local
network or area network (LAN), but it can be classified as a MAN,
extranet area WAN, or other type of network. At least one reference to
unit is usually
shielded from an external network should be included.
being accessed • Intranet: An intranet is a collection of interconnected
from the net if it computer networks managed by a single organization.
is not authorized.
They utilize IP-based tools, such as web browsers and file
transfer programs, and leverage the Internet Protocol. This
organization restricts access to the computer network to
only authorized users and blocks access for everyone else.
This network typically serves as an internal network for
an organization or business. Large computer networks can
typically support multiple internet servers to provide users
with accessible data.
• Internet: A selective Internetworking that consists of a
global network of public, academic, governmental, and
private networks connected primarily by the Advanced
Research Project Agency Network (ARPANET), created
by the U.S. The World Wide Web (WWW), which is also
housed by the Department of Defense and referred to as the
“Internet” to distinguish it from other generic Internetworks,
is another resource. Internet users and/or their service
providers use IP addresses that are sourced from address
registries that assign addresses.
Several major issues, including duplicate resource usage,
unmanaged networks, and isolated local area networks, have led
to the evolution of internetworking. Transmission issues between
completely separate offices or departments were caused by isolated
LANs. Due to resource duplication, each workplace or department
required a dedicated support staff member in addition to constant
hardware and code. Due to a lack of network management, there
 INTERNETWORKS 149
was no centralized approach to maintaining and debugging networks. Another way that
networks are typically connected is through the Link Layer of the networking model,
where businesses connect to one another. This layer is below the number of TCP/IP logical
interfaces and is hardware-centric. Network switches and network bridges enable this
kind of interconnection. Although this is sometimes mistakenly called “internetworking,”
the resulting system is really just a larger, single sub-network, and it doesn’t require
an internetworking protocol—similar to web protocol—to move between these devices.
Nevertheless, by segmenting the network into phases and using routers to logically
split the segment traffic, one electronic network can also be reborn into an associate
degree internetwork. The purpose of the Internet Protocol is to provide a somewhat
erratic packet service throughout the network. The network’s state is not maintained
by intermediary network components, as per the design. Rather, this task is assigned
to each communication session’s endpoints. Applications should use an appropriate
Transport Layer protocol, such as Transmission Control Protocol (TCP), which offers a
dependable stream, to transfer data accurately. For tasks like video streaming or voice
chat that require temporary service or don’t require dependable information delivery,
some applications use the simpler, connectionless User Datagram Protocol (UDP).

Did you get it?

1. What are the various challenges encountered in building and managing interconnected networks?
2. Can you explain the concept of a unit of internetworking and its role in establishing connections
and data exchange between devices?
150 COMPUTER NETWORKS

4.2. INTERNETWORK
ARCHITECTURE
→Learning Objectives

• Demonstrate an understanding of how the Application Layer facilitates end-user

applications’ interaction
• Illustrate the transport Layer’s role
• Explain the purpose and significance of the Internet Layer
• Explore various technologies and protocols operating at the Network Access Layer

An internetwork is a group of broadcast and packet-switching networks that work together

as a single, sizable network thanks to the use of routers, switches, and bridges, which
are intermediate networking devices. As a result, regardless of the network segment
to which they are connected, all users and devices can communicate. An internetwork
can be created by connecting various types of network technologies, as shown in Figure
4.5, through the use of routers and other networking devices.

Figure 4.5. Internetworking flow meter (Source: CertificationKits, https://www.certificationkits.

com/assets/images/stories/image16.jpg).
 INTERNETWORKS 151
Local Area Networks (LANs) and Wide the Internet Protocol Suite is a framework
Area Networks (WANs) are two categories that specifies a set of general design
for the various network types that combine guidelines and implementations of particular
to form an Internetwork. A LAN is a network networking protocols. The Internet Protocol
that is restricted to a small geographic area, Suite architectural model, also known as
like an office building or even a network the TCP/IP Model, is described in RFC
of computers and devices in a home. 1122. A host must implement the layered
There are numerous uses for a Local Area set of protocols that make up the Internet
Network (LAN), including resource sharing, protocol suite in order to communicate over
gaming, and teamwork. When a company the Internet.
owns a Local Area Network, everything
from switches to cables is part of the The uppermost layer is called the
network’s infrastructure. Although a local Application Layer, and it describes the
area network can connect users at speeds of higher-level protocols, like File Transfer
up to 10 Gbps, there is a distance restriction Protocol (FTP), Hyper Text Transfer Protocol
on the media. (HTTP), and Simple Mail Transfer Protocol
(SMTP), that are used by applications
Wide Area Networks (WAN) span large for network communication. Applications
geographic areas, including entire countries, can communicate end-to-end through the
cities, and even continents. Multiple Local transport layer.
Area Networks that are spread out over
several miles are connected by Wide Area There are two primary transport layer
Networks. An organization must purchase protocols:
WAN service from a service provider or • Transmission Control Protocol
carrier due to the long distance connectivity. (TCP); and
Wide Area Networks come in a variety of
• User Datagram Protocol (UDP).
forms, including packet-switched, circuit-
switched, and leased line networks. While TCP is a reliable connection-oriented
Local Area Networks are owned by the transport protocol that provides flow control,
organization itself and do not require any re-sequencing, and end-to-end reliability.
kind of connectivity service purchase, UDP is a connectionless (“datagram”)
Wide Area Networks are contracted by transport protocol. To transmit datagrams
businesses or organizations that need to (packets) from a host across network
connect multiple offices located across boundaries to the destination host identified
cities, countries, or even the entire world. by a network address (IP address) defined
by the Internet Protocol (IP), a set of
The TCP/IP Protocol Stack, also referred techniques, protocols, and specifications
to as the Internet Protocol Suite, is a are detailed in the Internet Layer. Each
standard set of communication protocols computer on the Internet needs a unique
that was created as a result of DARPA IP address for identification, whether it is
research and development for the ARPANET. a host computer using services or a server
This protocol suite was created in order to providing services. The current version of
establish a common means of communication the Internet Protocol used on the Internet
that would enable various network types is IP version 4 (IPv4). The 32-bit address
with various computer systems to speak specified by IPv4 allows for 4,294,967,296
with one another. In order to facilitate unique addresses; however, due to the rapid
computer communication over a network,
152 COMPUTER NETWORKS

growth of the Internet, this space is limited. The address space of

IP version 6 (IPv6), intended to replace IPv4, is 128 bits, significantly
larger than that of IPv4. The Network Access Layer, also known as
the Link Layer or the Media Access Layer, defines protocols and
Remember specifications that the host must follow to communicate with its
All Internet directly connected network.
transport
protocols use the
Internet Protocol a. How the TCP/IP Protocols Handle Data Communications
(IP) to carry data
from source host A set of actions is started when a user issues a command that
to destination makes use of the TCP/IP application layer protocol. The local
host. IP is a
system’s TCP/IP protocol stack processes the user’s command or
connectionless
or datagram message. Subsequently, the command or message is transmitted
internetwork through network media to the remote system’s protocols. Each
service, providing layer’s protocols on the sending host supplement the original data
no end-to- with new information. Every layer of the sending host’s protocols
end delivery
guarantees.
communicates with its peers on the receiving host. The TCP/IP
protocol stack is used by two hosts to communicate, as seen in
the figure below.

Figure 4.6. How information travels on internet? (Source: Medium.

com, https://miro.medium.com/v2/resize:fit:720/format:webp/1*id6bv--
a6s3oGk2uO0jnEg.jpeg).

The Internet, the most famous example of an Internetwork, is

a global network that emerged from connecting various computer
networks. It comprises both private and public networks of different
sizes and types, serving diverse purposes and utilizing a wide array
of technologies. The ARPANET, developed by the United States
Defense Advanced Research Projects Agency (DARPA) in the 1960s,
 INTERNETWORKS 153
laid the groundwork for the Internet. In the 1980s, the government-
sponsored military research project ARPANET became known to
the public, leading to the rapid expansion and interconnection of
various computer networks that eventually formed the Internet.

Did you know?

In 1977, ARPA demonstrated a three-way internetworking experiment, which linked a
mobile vehicle in PRNET with nodes in the ARPANET, and via SATNET, to nodes at UCL.
The X.25 protocol, on which public data networks were based in the 1970s and 1980s, was
supplemented by the X.75 protocol which enabled internetworking.

b. Key Internet Services, and technologies and protocols that enable

these services
The Internet of today is a widely dispersed global network that is
accessible through multiple channels from any location. The Internet
began as a defense research project and has since developed into
a global network used for a variety of applications, including social
networking, multimedia, and business/corporate communications.
Using a variety of technologies and protocols, the Internet provides
a wide range of services, some of which are covered below.
• World Wide Web: The Internet’s World Wide Web (WWW) is
a network of linked hypertext documents. A web browser
allows users to view and use hyperlinks to navigate between
web pages that may contain text, images, videos, and other
types of multimedia. One of the main features that attracts Keyword
people to the Internet is the vast number of websites, Uniform
millions of them, offering various kinds of content. The Resource
Hyper Text Transfer Protocol (HTTP), an application layer Locator,
colloquially termed
protocol used by web browsers acting as HTTP clients to a web address, is
request information from web servers hosting websites, a reference to a
supports the entire World Wide Web. A website address is web resource that
specifies its location
found using a Uniform Resource Locator (URL), and resources on a computer
and information are requested from web servers via HTTP network and a
via URLs. mechanism for
retrieving it.
• Email: Email, also known as electronic mail, is a way to send
and receive digital messages. One of the most widely used
services provided by the Internet, next to the World Wide
Web, is email. The email system, which enables message
exchanges between Internet users, is based on a number
of standards and protocols. Email transmission over Internet
Protocol (IP) networks is facilitated by the Simple Mail
Transfer Protocol (SMTP), an Internet standard. Email clients,
154 COMPUTER NETWORKS

or end users, typically use Post point, the application or data is displayed
Office Protocol (POP) or Internet in a way that the user can comprehend
Message Access Protocol (IMAP) visually.
to access their mailbox accounts
on mail servers, while email servers Functions of the application layer:
use SMTP to send and receive mail • Ensures that the receiving device
messages. is identified, can be reached, and
• Internet Infrastructure: An Internet is ready to accept data.
Service Provider (ISP) is a company • Enables, if appropriate,
that connects individual users or authentication to occur between
corporate private networks to the devices for an extra layer of security.
Internet. ISPs serve a large number
• Makes sure necessary
of users through a variety of
communication interfaces exist.
physical media, including DSL and
For example, is there an Ethernet
wireless connections. These ISPs
or Wi-Fi interface in the sender’s
are referred to as local ISPs in an
computer?
Internet infrastructure. While local
ISPs can peer with one another, • Ensures agreement at both ends
they typically peer with regional about error recovery procedures,
ISPs, which link several local ISPs data integrity, and privacy.
together. Regional ISPs then peer • Determines protocol and data
with the larger provider known syntax rules at the application level.
as the Network Service Provider • Presents the data on the receiving
(NSP). This may not, however, be end to the user application.
taken to be the actual Internet
infrastructure, as different peering, Since the application layer in the OSI
connections, and other complexities model is the one nearest to the end user,
make it challenging to graph the both the user and the application layer can
entirety of the Internet. communicate with the software program
directly. Applications at the application layer
are client-server based.
4.2.1. Application Layer
The application layer includes the
In order to enable efficient communication
following functions:
between an application program and
another application program on a network, • Identifying communication
the application layer offers services to the partners: The application layer
program. It is not appropriate to think of identifies the availability of
the application layer as an application in communication partners for an
the sense that most people do. Rather, an application with data to transmit.
application’s application layer is a part that
• Determining resource availability:
manages how the application communicates
The application layer determines
with other devices. The remainder of the
whether sufficient network
application is hidden from the transmission
resources are available for the
process by this abstraction layer service. To
requested communication.
finish its task, the application layer depends
on every layer that comes before it. At this • Synchronizing communication: All
 INTERNETWORKS 155
the communications occur between the applications require
cooperation, which is managed by an application layer.
Services of Application Layers
• Network Virtual terminal: An application layer allows a
user to log on to a remote host. To do so, the application
creates a software emulation of a terminal at the remote
host. The user’s computer talks to the software terminal,
which in turn, talks to the host. The remote host thinks
that it is communicating with one of its own terminals, so
it allows the user to log on.
• File Transfer, Access, and Management (FTAM): An
application allows a user to access files in a remote computer,
to retrieve files from a computer and to manage files in a
remote computer. FTAM defines a hierarchical virtual file
in terms of file structure, file attributes and the kind of
operations performed on the files and their attributes.
• Addressing: To obtain communication between client and
server, there is a need for addressing. When a client made
a request to the server, the request contains the server
address and its own address. The server response to the
client request, the request contains the destination address,
i.e., client address. To achieve this kind of addressing, DNS
is used. Keyword
• Mail Services: An application layer provides Email forwarding File Transfer
Protocol (FTP) is
and storage. a standard network
• Directory Services: An application contains a distributed protocol used
for the transfer
database that provides access for global information about of computer files
various objects and services. between a client
and server on a
• Authentication: It authenticates the sender or receiver’s computer network.
message or both.
A network diagram is usually used to guide the design of a
computer network. This topology, which has seven layers of network
functionality, is based on the Open Systems Interconnection model
(OSI). Known as layer seven in OSI, the application layer is the
highest level. The Hypertext Transfer Protocol (HTTP), email, and
File Transfer Protocol (FTP) are among the communication protocols
that are managed by the network layer. These applications allow
email clients and web browsers to share files over a network.ile
transfer protocol (FTP) are among the communication protocols that
are managed by the network layer. These applications allow email
clients and web browsers to share files over a network.
A physical cable is the first step in a network, and a computer
program is the last. In the course of a typical communication
156 COMPUTER NETWORKS

transfer, data is forced through multiple messages falls under the purview of
levels of transformations. To guarantee the lowest tier of the OSI model. The
appropriate data quality and security presentation layer, which manages the
regulations are followed, this procedure is message’s format and encryption, receives
required. Software protocols can now be messages from the application layer.
more flexible thanks to the creation of the
OSI model. There are seven layers in the OSI
model. The transport, data link, network, 4.2.2 Transport Layer
session, presentation, and application layers Data that must be transported beyond
are these. The layers communicate over the confines of an address space must go
a network by working in a hierarchical through a transport layer. Unstructured
structure. Standard transmission control byte sequences make up the data being
protocol (TCP/IP) is the Internet’s primary transported from the perspective of the
communication language and is used by transport layer. A concrete transport
the majority of computer networks today. mechanism is used to implement the
The OSI model serves as a guide for how transport layer. The transport layer’s job
network hardware and software should is to provide a uniform interface to the
communicate in this network protocol. In presentation layer while abstracting from
the network topology, every layer in the platform-specific transport mechanisms.
model has a distinct role to play. Generally speaking, the transport layer is
just an operating system’s wrapper around
The hierarchical chaining of responsibility
network functions. Just like when the
is the foundation of the OSI model. These
bank scenario is introduced, an analysis
guidelines compel every layer to transmit
of requirements shows the features of the
messages through the topology by adhering
transport layer that serve as the foundation
to predetermined handoff procedures. The
for a design:
presentation layer, the next tier down the
hierarchy chain, has handoff rules that the 1. The transport layer guarantees
application layer must abide by. The next a reliable end-to-end connection
level in the chain usually requires certain between precisely two address
size requirements, format, and encryption spaces.
before accepting a message. One common 2. Data can be sent bi-directionally
communication protocol on computers and in the form of unstructured byte
mobile devices is instant messaging, or IM. sequences of any length.
Software that makes use of the application
layer protocol is instant messaging (IM). 3. Different transport mechanisms
Instant messaging software is made possible should be supported.
for businesses by the IM standard protocol.
AOL Instant Messenger, Yahoo Mail, and 4.2.2.1. Transport Layer Links Address
MSN Messenger are a few of these products. Spaces
At this layer, every program communicates The details of the underlying transport
over the network using the OSI standards. mechanism are hidden above the transport
One email program that makes use of layer, which ensures reliable end-to-end
the application layer is Microsoft Outlook. connections. Several transport protocols,
The task of converting software program such as TCP, currently offer reliable
messages into comprehensible computer connections. It becomes more challenging to
 INTERNETWORKS 157
guarantee appropriate semantics above the uneven application of the transport layer.
transport layer when an unreliable transport Based on the matching communication
mechanism is used at that layer. Requiring endpoints, clients and servers define a
the transport layer to connect precisely communications channel. An address
two address spaces aims to simplify the designating the communication endpoints
middleware. Managing transport channels is a sign of the client and server’s consent.
between different address spaces would Usually, a network address referring to the
be problematic without this restriction. actual transport mechanism is included in
Therefore, μORB only permits an application’s a transport layer address. Address formats
objects to be split between exactly two and details can vary amongst transport
address spaces. mechanisms. Thus, there isn’t a format that
is generally accepted.
The transport layer views data as
unstructured, variable-length byte sequences
exchanged between a client and a server. 4.2.2.3. Classes Buffer, Address, and
While the transport layer is not expected Transport Model Transport Layer
to continuously transmit media, multimedia
To model the transport layer, three classes—
applications require this capability. However,
Buffer, Address, and Transport—are
it is expected to support various transport
introduced. The storage region connected
mechanisms.
to a specific memory chunk is represented
The following use case outlines the type by the class buffer. The unstructured byte
of modeling needed at the class level: sequences that are transferred between
address spaces via a transport mechanism
1. The server establishes a are contained in instances of this class.
communication end point and waits There are methods available for configuring
for a connection setup request. and managing a memory chunk through the
2. The client creates a communication class Buffer’s interface. A specific transport
end point and connects it to the mechanism’s address is represented by
server. the class Address, and the transport
3. The client sends data and waits mechanism itself is represented by the
for a response. class Transport. This address serves as the
transport mechanism’s factory. Since they
4. The server receives the data of the merely serve as the interface for the actual
client and then sends a response. transport mechanism, the two classes are
5. After the client receives the abstract in and of themselves. As a result,
response from the server, both close each transport mechanism requires the
their communication end points definition of two concrete classes: one for
the address and another for the transport
4.2.2.2. Applications Scenario of the mechanism itself. The abstract base classes
Transport Layer Address and Transport are the ancestors
of these classes. A TCP-based transport
Keep in mind that the terms “client” and mechanism in UML notation is displayed
“server” do not refer to actual objects in Figure 4.7.
in this use case. Instead, they discuss
the roles that are played in interactions
with the transport layer, suggesting an
158 COMPUTER NETWORKS

Figure 4.7. UML class diagram of transport layer (Source: ScienceDirect,

https://ars.els-cdn.com/content/image/3-s2.0-B978155860648750005X
-f04-02-9781558606487.gif).

4.2.2.4. Responsibilities of a Transport Layer

Keyword The second layer in the TCP/IP model is called the Transport Layer.
User Datagram In order to send messages to a host, it is an end-to-end layer. It
Protocol (UDP) is called an end-to-end layer because, in order to reliably deliver
is an alternative
communications the services, it establishes a point-to-point connection between the
protocol to TCP source host and destination host as opposed to a hop-to-hop one.
used primarily A segment is the data encapsulation unit used in the Transport
for establishing
low-latency and Layer. TCP (Transmission Control Protocol), UDP (User Datagram
loss-tolerating Protocol), DCCP (Datagram Congestion Control Protocol), and other
connections common protocols are utilized by the Transport Layer to improve
between
applications on the its functionality.
internet.
Various responsibilities of a Transport Layer:
• Process-to-process delivery: Just as the Data Link Layer
needs the MAC address (a 48-bit address found inside each
host machine’s Network Interface Card) of the source and
destination hosts in order to correctly deliver a frame, and
the Network Layer needs the IP address in order to route
packets appropriately, the Transport Layer also needs a
port number in order to correctly deliver data segments to
the appropriate process among the various processes that
are running on a given host. Any client-server program
can be uniquely identified by its port number, which is a
16-bit address.
• End-to-end Connection between hosts: TCP and UDP are
primarily used by the transport layer to create this connection,
which is also within its purview. TCP is a secure, connection-
oriented protocol that builds a strong connection between
two end hosts through a handshake protocol. TCP is used
 INTERNETWORKS 159
in many different applications and congestion control to eliminate
guarantees dependable message congestion in a network after it has
delivery. On the other hand, UDP already occurred. AIMD—additive
guarantees best-effort delivery and increase multiplicative decrease—
is a stateless, unstable protocol. and the leaky bucket technique are
It works well for applications like two of TCP’s congestion control
video conferences that need to send mechanisms.
large amounts of data and don’t • Data integrity and error correction:
care too much about flow or error The transport layer uses error
control. Frequently, multicasting detection codes, computes
protocols make use of it. checksums, verifies that the
• Multiplexing and demultiplexing: received data is not corrupted,
Multiplexing enables the use of uses the ACK and NACK services
several applications simultaneously to inform the sender whether the
over a network while a host data has arrived, and verifies the
is operating. This mechanism, integrity of the data. It also checks
provided by the transport layer, for errors in the messages that
allows us to send packet streams come from the application layer.
from multiple applications over a • Flow control: In the TCP/IP model,
network at the same time. After the transport layer acts as a conduit
adding the appropriate headers, for data between the higher and
the transport layer transfers these lower layers. TCP also employs
packets—which are identified by flow control techniques to prevent
their port numbers—from various data loss from a fast sender and
processes to the network layer. slow receiver. It utilizes the sliding
Demultiplexing is also necessary window protocol, which is executed
at the receiving end in order to by the receiver informing the sender
receive data originating from of the maximum amount of data
different processes. The appropriate it can receive by sending back a
process operating on the recipient’s window.
computer receives the data
segments that transport receives
from the network layer. 4.2.3. Internet Layer
• Controlling congestion: Congestion The Internet layer is in charge of organizing
occurs when there are too many data that needs to be sent into IP datagrams,
sources attempting to send data which are packetized data packets. These
over a network, causing the will have the data’s source and destination
router buffers to overflow and addresses listed. The IP datagram routing
packet loss. Consequently, the is another function of this layer. Address
congestion is exacerbated by the Resolution Protocol (ARP), Reverse Address
sources’ retransmission of packets. Resolution Protocol (RARP), Internet Group
Congestion control is provided in Management Protocol (IGMP), Internet
this scenario by the Transport layer Protocol (IP), and Internet Control Message
in various ways. It makes use of Protocol (ICMP) are the primary protocols
both closed-loop and open-loop present at the Internet layer.
160 COMPUTER NETWORKS

4.2.3.1. Static IP is a particular IP address that is set aside

for the exclusive use of a particular DHCP
An IP address that has been manually client within the scope of DHCP. In controls
configured for a device is called a static networks, lease reservations are preferred to
IP (also called a fixed IP address; see view dynamic leases. Sending and receiving data
from 10:20 to 13:20). Because it remains to other devices on the same network can
constant without user intervention, the IP depend on a controller’s IP address. It will
address is known as static. An IP address be simpler to configure and maintain the
that matches the range selected for the system if the IP address remains constant.
task or provided by the IT department must In order to utilize lease reservations, you
be manually assigned when configuring a will require knowledge of the controller’s
controller. It is necessary to manually assign MAC address. In order for the person in
a unique IP address to every device on charge of the DHCP server to make sure
the network. Static IP addresses have a every controller receives the IP address
significant drawback over dynamic ones: it is meant to have you will need to give
manually configuring the devices is required. them this.
This is usually done device-by-device.

4.2.3.2. DHCP
4.2.4. Network Access Layer
The Data Link and Physical layers of the
A client/server protocol called Dynamic Host OSI reference model align with the Network
Configuration Protocol gives an IP host its Access layer of the TCP/IP model. It outlines
IP address and other relevant configuration the hardware and protocols needed to link
data, such as the subnet mask and default a host to a real network and transfer data
gateway, automatically. With DHCP, IP between them. In order to be delivered
addresses and other network configuration within the physical network, packets from
data can be distributed and updated the Internet layer are routed through the
automatically. When a client submits a Network Access layer. The destination may
request, a DHCP server usually replies with be the network host itself, another host, or a
an address that is within the DHCP scope. router for additional forwarding. Therefore,
If they are set up to do so, DHCP servers the Network Access layer is restricted to the
can offer more network-related information physical layer boundary, which is typically
in addition to IP addresses. A valid range of defined by a layer 3 device like a router,
IP addresses that can be leased or assigned while the Internet layer has a view of the
to client computers on a specific subnet is entire Internetwork.
known as a DHCP scope. A scope is set up
in a DHCP server to ascertain the address There are numerous protocols that make
pool of IPs that the server can offer to up the Network Access layer. The most
DHCP clients. The DHCP server can assign widely used protocols when the physical
an address to a client, extend the lease network is a local area network (LAN)
term, recognize when the address is not in are Ethernet and its numerous variants.
use, and reclaim it when an address has a Conversely, protocols like Frame Relay and
dynamic lease. Point-to-Point Protocol (PPP) are frequently
used when the physical network is a Wide
A DHCP reservation, on the other hand, Area Network (WAN).
is a permanent IP address assignment. It
 INTERNETWORKS 161
• The Network Access layer uses a physical address to identify
hosts and to deliver data.
• The Network Access layer PDU is called a frame. It contains
the IP packet as well as a protocol header and trailer from
this layer.
• The Network Access layer header and trailer are only relevant
in the physical network. When a router receives a frame, it
strips off the header and trailer and adds a new header and
trailer before sending it out to the next physical network
towards the destination.
The TCP/IP protocol hierarchy starts at the Network Access Keyword
layer. Because it is well hidden by the more well-known higher-level Network
protocols like SMTP, HTTP, and FTP, as well as mid-level protocols like interface
card (NIC)
IP, TCP, and UDP, it is frequently disregarded by users. Encapsulating is a hardware
IP datagrams into network-transmitted frames and translating IP component
addresses into physical hardware addresses are among the tasks without which a
computer cannot
carried out at the network access layer. Software programs and be connected over a
drivers specific to particular hardware components handle a large network.
portion of the work at the network access layer. Most of the time,
configuration is as simple as choosing the right driver to load and
TCP/IP as the protocol to use. Many computers have this driver
software pre-installed and set up, or they have “plug-and-play” apps
that allow them to configure themselves automatically.
Configuring a Windows NT system to use a 3Com Etherlink
III network interface card (NIC) with Ethernet is a good example
of network access layer configuration. In a typical scenario, the
operating system would identify the NIC during installation. If this
doesn’t happen, or if the card is added later, installing the card and
picking the driver by selecting the card’s manufacturer and model
from a list of available drivers would be the main steps in the
installation process. Certain cards may also need minimal hardware
configuration, including deciding on an I/O port, setting the IRQ,
and maybe utilizing the card’s Twisted Pair or Coax connector. Upon
selecting and configuring the appropriate driver for the card, the
subsequent action involves choosing TCP/IP from a list of available
protocols to be utilized on the interface. The network access layer
configuration is finished after this. Numerous configuration programs
proceed to configure DNS servers, IP addresses, netmasks, default
gateways, and other parameters that are actually related to other
layers and protocols automatically.
It’s true that some network access layer protocols require a lot
of configuration. Understanding the operation of the more common
ones is beneficial. Network managers should know about serial point-
to-point protocols like SLIP and PPP if they’ll be handling remote
162 COMPUTER NETWORKS

access over serial communication lines, such as ISDN lines or POTS modems. Being
aware of remote authentication protocols like TACACS and RADIUS is helpful if they’ll
be used on Network Access Server hardware, such as modem stacks or terminal servers.

Did you get it?

1. What are the key Internet services, and what technologies and protocols enable these services?
2. Can you explain the classes Buffer, Address, and Transport model within the transport layer?
 INTERNETWORKS 163

FOCUS ON CAREERS
NETWORK ARCHITECT
A network architect is responsible for designing and building data communication
networks. The networks they work on can range from smaller local area networks
(LANs) to larger wide area networks (WANs), along with intranets. These professionals
also manage the infrastructure of computer networks to ensure they operate effectively,
including the installation of routers, cables, modems, and other necessary hardware
and software. Network architects can find employment at organizations in numerous
industries, including telecommunications, finance, insurance, and computer systems
design services. This role is typically a senior IT position within an organization, often
working alongside a team of IT specialists and communicating with management staff.
These professionals use their understanding of the organization’s business goals to
develop network systems that meet them. If you are thinking about becoming a network
architect, learning about the typical requirements for this career may help you assess
whether this profession is a good fit for you:

Education and Training

To pursue a career as a network engineer, one typically needs to obtain at least a bachelor’s
degree in a relevant field. For instance, a degree in computer science, information systems,
engineering, or mathematics may be considered. Coursework in computer-related fields
provides individuals with hands-on experience, relevant skills, and knowledge necessary
to carry out the typical job responsibilities of a network architect. Additionally, individuals
learn about various technologies, software, and programs utilized in the industry. A
bachelor’s degree usually requires around four years to complete. Some employers or
higher-level positions may mandate candidates with more advanced degrees, such as a
master’s or doctorate. One possibility is pursuing a Master of Business Administration
degree in information systems. An MBA program encompasses both computer- and
business-related coursework, including project management. An advanced degree could
potentially facilitate career progression and showcase extensive expertise in the field.
A master’s program typically takes about two years to finish, while a doctorate may
take four to six years.

Certifications
Network architects have several certifications they can obtain, often through specific
software and product vendors. Many of these certifications have different levels that you
can earn as you continue to advance your IT skills and knowledge. The certification you
pursue may vary based on the types of systems or software used or required by your
employer. You can research job openings in your field to understand the certifications
164 COMPUTER NETWORKS

sought by employers. Some examples of common certifications pursued by network

architects include:
• Cisco Certified Network Professional (CCNP)
• Cisco Certified Architect (CCAr)
• CompTIA Network+
• ITIL Master
• Microsoft Technology Associate (MTA)
• Red Hat Certified Architect (RHCA)
• Salesforce Certified Technical Architect
• VMware Certified Professional (VCP)
• Zachman Certified – Enterprise Architect

Work Experience
When hiring for network architect roles, employers typically seek candidates with at
least five years of professional experience working with IT systems. The amount of
experience sought may vary by employer or role, with some requiring at least 10 years.
You may gain relevant experience through internships as a student or entry-level jobs
in the field. For example, individuals may work as network engineers, administrators,
computer systems analysts, or database administrators. Demonstrating to potential
employers that you have relevant work experience can prove that you have the skills
and knowledge needed to perform the job effectively. When possible, you can also share
any achievements from your prior roles to set yourself apart from other candidates.

Skills
Network architects need numerous technical skills to perform their job effectively, such
as those related to computer operating systems, wireless systems, and security. They
typically learn such skills and knowledge through their education, training programs,
or on-the-job experience. Network architects also benefit from having several other
skills, including:
• Analytical thinking: Network architects must examine networks to identify
issues and develop potential solutions. They also use this skill to determine
the appropriate hardware, software, and other equipment to utilize based on
their organizations’ resources and needs.
• Attention to detail: When designing complex data communication networks,
network architects must consider numerous details. Having an attention to detail
helps ensure they create comprehensive designs with accuracy and networks
that function effectively.
• Communication: Network architects work with other IT professionals, and
strong verbal and written communication and interpersonal skills can ensure
 INTERNETWORKS 165

they convey information effectively and accurately. They may also use these
skills to present plans and provide networking guidance to management staff.
• Leadership: Network architects may need to lead a team of IT specialists or
computer engineers through the network implementation process or other
necessary tasks. Leadership skills enable them to assign responsibilities and
maintain the proper functioning of the network.
• Organizational skills: Depending on their job, network architects may need to
manage numerous networks within an organization and keep these different
systems organized to ensure they work well together. Organizational skills,
such as time management and strategic planning, also help keep projects on
schedule and budget.
166 COMPUTER NETWORKS

4.3. IP ADDRESSING
AND ARCHITECTURE
→Learning Objectives

• Define what an IP (Internet Protocol) address is and its significance in network

communication.
• Describe the various types of IP addresses used in networking, such as unicast,
multicast, and broadcast addresses.
• Explain the concept of website IP addresses and their role in hosting web
content on the internet.
• Classify IP addresses based on their operational characteristics, such as static
and dynamic IP addresses.
• Understand the basic functioning of IP addresses in routing data packets
between devices on a network.
• Define static IP addresses and their role in network configuration, particularly
in scenarios where consistent, unchanging addresses are necessary.
• Identify tools and techniques used to track and record IP addresses, such as
network monitoring software and log files.
• Differentiate between public and private IP addresses in terms of their scope,
accessibility, and usage.

An Internet Protocol address is represented by an IP address. A special address that

serves as the device’s network identification. It resembles a set of guidelines controlling
the format of data transferred via a local network or the Internet. An IP address
facilitates the Internet’s ability to discern between various routers, PCs, and WebPages.
It facilitates better visual communication between the source and the destination and
acts as a unique machine identifier within a particular network.
IP address structure: IP addresses are displayed as a set of four digits- the default
address may be 192.158.1.38. Each number on the set may range from 0 to 255. Therefore,
the total IP address range ranges from 0.0.0.0 to 255.255.255.255.
IP address is basically divided into two parts: X1. X2. X3. X4
1. [X1. X2. X3] is the Network ID
 INTERNETWORKS 167
2. [X4] is the Host ID (IPSec) is required in this. It appears
1. Network ID: This portion of the to be more than enough to support
IP address on the left-hand side the trillions of internet-connected
designates the particular network devices that are currently in use
that the device is connected to. or will be in the future because
When a device is connected to a it permits 3.4 x 10^38 unique IP
typical home network and its IP addresses.
address is 192.168.1.32, it is called
192.168.1. The network ID will be 4.3.2. IP Address Types
one component of the address.
IP addresses come in four varieties: public,
Since it is usual to fill in the final
private, fixed, and dynamic. These include
digit that is not zero, we can state
their local network location, which should
that the network ID of the device
be used within the network, and their public
is 192.168.1.0.
IP address, which should be used online.
2. Hosting ID: The portion of the IP
address that the network ID did not 1. Public IP address: An IP address
take is the host ID. It identifies a that is assigned by different servers
particular device in that network and devices is known as a public IP
(called the “host” in the TCP/ address. When you connect these
IP world). Using the IP address devices to your internet connection,
192.168.1.32 as an example again, this IP address is displayed on our
the host ID will be 32, making it homepage. The reason for this
the only host ID on the 192.168.1.0 explanation is that not everyone is
network. familiar with IP communication. Our
objective is to simplify the process
of obtaining necessary information
4.3.1. Version of IP address for all individuals. Some even refer
Currently there are 2 versions of IP to this IP address as their external
addresses are in use i.e IPV4 and IPV6 one. A public IP address that can be
accessed over the Internet is unique
1. IPv4 (Internet Protocol Version 4):
and is assigned to a computer
It is the original Internet Protocol
device, similar to a postal address
address version. IPv4 addresses
used for mail delivery to your home.
have a 32-bit length. Regarding
Servers that will use the public IP
network security, Internet Protocol
address include web servers, email
Security (IPSec) is optional in this
servers, and any other server device
case. Despite having 4,294,967,296
with direct Internet access. The
addresses, there is still a shortage
Internet Protocol address is unique
of network addresses due to the
to a single device and is globally
rapidly growing number of network
unique.
and virtual devices being used.
2. Private IP address: The IP address
2. IPv6 (Internet Protocol Version 6):
of every device connected to your
It is the most recent iteration of the
network over the Internet is unique.
IP address. IPv6 addresses have a
This includes not only PCs, iPhones,
bit count of 128. In terms of network
and iPads but also any Bluetooth-
security, Internet Protocol Security
168 COMPUTER NETWORKS

capable devices like smart TVs, can be inconvenient, dynamic IP

printers, and speakers. With the addresses allow for easy and cost-
expansion of the Internet of Things, free navigation with the correct
your home likely has more private software.
IP addresses now. These devices
need to be uniquely identified 4.3.3. Types of Website IP
by your router, and most devices
require some form of communication address
protocol. To differentiate each Website IP addresses come in two types
device within the network, your – Dedicated IP Address and Shared IP
router assigns private IP addresses. Address. Let’s delve into both.
3. Static IP Address: An IP address • Dedicated IP address: Each website
that is static is not valid. In is assigned a specific dedicated
contrast, a dynamic IP address that IP address that is unique to that
is subject to change will be supplied domain. This means no other
by the Dynamic Host Configuration domain shares the same address.
Protocol (DHCP) server. Although it There are several benefits to having
can be modified as part of routine a dedicated IP address, such as
network administration, the static increased speed and enhanced
IP address is not dynamic. Once security, especially during high
assigned, static IP addresses do not traffic periods. However, dedicated
change over time and are therefore IPs are generally more costly
incompatible. Additionally, you can compared to shared IPs.
learn more about the device with
the aid of this kind of IP. • Shared IP address: A shared IP
address is one that is not unique
4. Dynamic IP Address: It implies and is shared by multiple domains.
ongoing change. A dynamic IP For most users, a shared IP address
address is one that is not constant suffices as common configurations
and varies over time. You might do not require a dedicated IP
have a dynamic IP address if you address.
have active cable or DSL service.
Because they are less expensive,
Internet service providers give 4.3.4. IP Address Classification
their clients dynamic IP addresses. Based on Operational
Your IP address is pulled from the Characteristics
address pool and assigned to you,
as opposed to being assigned a Based on operational characteristics, IP
single permanent IP address. That addresses are classified as follows:
number is returned to the pool Broadcast addressing: Broadcasting
and given a new number after a involves sending audio or video data
few days, weeks, or occasionally across a network. A broadcast packet
even months. If an ISP does offer is simultaneously sent to all users on a
a static IP address to a customer local network without the need to specify
who resides there, it’s usually individual recipients. Users can choose to
at a higher cost. Though they open, read, follow instructions, or delete
 INTERNETWORKS 169
these packets. This service is available by a (DHCP) server and is not fixed, unlike
in IPv4 and typically uses the IP address a static address. The most prevalent kind
255.255.255.255. of internet protocol addresses is dynamic
IP addresses. Only a limited period of time
• Unicast addressing: This type of is allowed for dynamic IP addresses to
address identifies a specific node be active before they expire. Either the
on the network. Unicast refers computer will ask for a new lease on its
to the one-to-one transmission own, or it might get a new IP address.
of data between network nodes,
making it the most commonly
used IP addressing type. Both data
transmission and reception are Remember
possible with unicast addressing, An IP address uniquely identifies a device on a
available in both IPv4 and IPv6. network. You’ve seen these addresses before; they
• Multicast IP addresses: These IP look something like 192.168.1.34.
addresses mainly help to establish An IP address is always a set of four numbers like
one-to-many communication. that. Each number can range from 0 to 255. So,
Multicast IP routing protocols are the full IP addressing range goes from 0.0.0.0 to
used to distribute data to multiple 255.255.255.255.
recipients. The class D addresses
(224.0.0.0 to 239.255.255.255) define Since an IP address is specific to the
the multicast group. computer or user to whom it is assigned, it
• Anycast addressing: A packet is can be likened to a Social Security Number
not sent to every network receiver (SSN). Routers can determine where they
when using anycast addressing for are sending data over the internet by using
data. A data packet is sent to the these numbers. Additionally, they verify that
nearest address when it is assigned the messages are reaching the appropriate
to an anycast address. devices. A router requires an IP address
in order to deliver to the requested web
4.3.5. IP Addresses Work address, much like the post office requires
a mailing address to deliver a package.
Computers can send and receive data
over the internet with the help of an IP
address. The majority of IP addresses are 4.3.5.1. The Subnet Mask
just numbers, but as more people use Thus, how does your device distinguish
the internet, some addresses now include between the network ID and the host ID?
letters. IP addresses come in four varieties: They do this by using a second number that
static, dynamic, private, and public. Static is always associated with an IP address.
and dynamic indicate permanence, while We refer to that figure as the subnet mask.
public and private indicate the network’s
location—private is used inside a network, On most simple networks (like the ones
whereas public is used outside of one. in homes or small businesses), you’ll see
subnet masks like 255.255.255.0, where
A static IP address was generated by all four numbers are either 255 or 0. The
hand rather than assigned. Moreover, a position of the changes from 255 to 0
dynamic IP address is one that is assigned indicate the division between the network
170 COMPUTER NETWORKS

and host ID. The 255s “mask out” the network ID from the equation.
IP addressing details are displayed together with a default gateway
address in addition to the IP address itself and the corresponding
subnet mask. This address may be known by a different name on
the platform you are using. It may be referred to as the “gateway,”
“router address,” “default route,” or simply “router.” These all refer
to the same thing. When a device is supposed to send network
data to a network other than the one it is currently connected to
(one with a different network ID), it sends the data to the default
IP address.
The simplest example of this is found in a typical home network
A router that is linked to the internet via a modem is probably
present in a home network that has several devices connected
to it. Your internet provider may provide a modem/router combo
unit that includes the router, or it may be a stand-alone device.
Passing (or routing) traffic back and forth between the computers
and devices on your network and the more publicly facing devices
on the internet is what the router does. Let’s say you open your
browser and go to www.howtogeek.com. Your computer contacts
the IP address of our website with a request. Your PC sends traffic
to your router, which acts as a gateway, and your router then
forwards the request to our server because our servers are on the
Keyword internet rather than your home network. You see our website load
in your browser when the server sends the appropriate data back
Router is a
networking device to your router, which subsequently routes it to the device that
that forwards data made the request.
packets between
computer networks. Typically, routers are configured by default to have their private
IP address (their address on the local network) as the first host
ID. So, for example, on a home network that uses 192.168.1.0 for a
network ID, the router is usually going to be 192.168.1.1. Of course,
like most things, you can configure that to be something different
if you want.

4.3.5.2. DNS Servers

There’s one final piece of information you’ll see assigned alongside
a device’s IP address, subnet mask, and default gateway address:
the addresses of one or two default Domain Name System (DNS)
servers. We humans work much better with names than numerical
addresses. Typing www.howtogeek.com into your browser’s address
bar is much easier than remembering and typing our site’s IP address.
DNS searches for human-readable information, such as website
names, and converts them to IP addresses. It functions somewhat
 INTERNETWORKS 171
like a phone book. DNS accomplishes this would rather have a static address. Having
by keeping track of everything on a network a static IP address has benefits in some
of connected DNS servers throughout the circumstances. For instance, this kind of
internet. For the purpose of sending their IP address is frequently preferred by those
queries, your devices must be aware of the who play online games with distant players.
addresses of DNS servers. The hosting server recognizes them when
they join back into a game, restoring their
The DNS server IP addresses are score, position in the game, and other
frequently the same as the default gateway settings appropriately. A computer cookie
address on a small or home network. When that can be removed is less dependable than
devices send DNS queries to your router, it a static IP address. Every website has a
routes them to the DNS servers that your static IP address, sometimes referred to as a
router is set up to use. You can modify Uniform Resource Locator or URL. A website
those to use different DNS servers if you’d wouldn’t have a permanent address without
like. By default, these are typically the DNS a static address, which would force users
servers that your ISP offers. to “follow it” around the Internet and make
bookmarks meaningless. This brings up the
4.3.6. Static IP Address most frequent reason for which someone
chooses this kind of IP address: to host a
An Internet Protocol (IP) address is a website. One can create a website and serve
unique numerical address that is assigned as the server, hosting it on the Internet
to computers when they connect to the just like any other website, provided they
Internet. Every computer on the Internet has have a static IP and 24/7 broadband access.
a unique address, which allows computers The owner of a static IP can also host an
to send and receive information in order email server and operate other services,
to communicate with one another. All that such as forums or webcasts. Even though
a static IP address is is a “permanent” third parties offer all of these services, some
address that is linked to a single computer people would rather take on the role of
for a considerable amount of time. This is host. If the computer will be connecting
not the same as a dynamic IP address, which to a Virtual Private Network (VPN) or
is given out randomly at the beginning of enabling remote access through programs
every session and typically varies from one like RadminTM or PCAnywhereTM, then a
to the next. static IP is also recommended. Provided that
A person’s Internet Service Provider (ISP) the remote computer in question has a static
assigns their IP address. A sizable block of IP IP address, software that allows remote
addresses that fall inside a particular range access only needs to be configured once. In
are assigned to each ISP. When clients log on the event that its address changes, the user
to the Internet, the ISP server automatically will need to identify the remote IP address
assigns available IP addresses within that and update the software accordingly.
range to them. The dynamic IP address that Unless a client specifically requests a
was temporarily assigned returns to the static IP address, ISPs will automatically
pool for reassignment when an individual assign a dynamic IP address to them. Static
disables their Internet connection. Most IP addressing is typically more costly and
people find that dynamically assigned IP perceived as less private by some. Since IP
addresses work well, but some people addresses are routinely logged by websites,
172 COMPUTER NETWORKS

an address that never changes gives data Websites automatically track and log IP
mining companies easy access to a fairly addresses, storing the numerical addresses
comprehensive profile of people’s online in server logs, for all of the aforementioned
browsing habits. reasons and more. Every request made by
the IP address is noted and has a time
stamp attached. Logs are regularly cleared
4.3.7. Track and Record IP of older data to make room for updated
Addresses information. The duration that a website
An IP address, also known as an Internet retains IP logs varies, is set by the site
Protocol address, is a special number that administrator, and depends on numerous
is given to a computer when it connects variables. In order to determine which pages
to the Internet. Internet Service Provider are the most popular, websites also monitor
(ISP) records can be used to trace an IP and log IP addresses. To boost website
address back to a specific person. When traffic, the website can expand on its most
users navigate the pages of nearly any popular pages. Leaks in traffic can also be
website on the World Wide Web (Web), found by tracking IP addresses throughout
IP addresses are tracked and recorded. the website.
This is primarily due to two factors: site IP addresses are typically referred to
enhancement and security. Every website as “anonymous” data in website policies.
has a server to host it on. A request for a Nonetheless, websites can still associate
webpage at the specified location is sent identities with IP addresses in a variety of
by the user’s browser to the server when ways—even if the address is dynamic and
they click to access a website. The webpage changes with every Web session—thanks
is sent back to the IP address that made to computer cookies. Additionally, a lot
the request by the server. After that, the of websites have “Web bugs,” which are
visitor’s computer screen loads the page. a few pixels connected to an advertising
The speed at which pages load for company that can track and record IP
visitors may decrease if the server is addresses as they move from one website to
overloaded with requests. When there are another on the Internet, covertly assembling
too many requests being made at once, comprehensive surfing profiles of people
the server will “crash” or go down, making over months or years. For a website to
the website momentarily unavailable. track and profile a visitor, registration is
This might be classified as a Denial of not required.
Service (DoS) attack. An attack is known If cookies are enabled in the web
as a Distributed Denial of Service (DDoS) browser, then every visit to a website is
attack if it originates from a botnet, which usually time-stamped and saved to a cookie.
is a collection of compromised computers. Every page and link that is viewed on the
A malevolent cybercriminal may attempt website is typically added to the cookie
to obtain data from password-protected (along with the server logs). Cookies are
databases containing credit card numbers often kept, but server logs are cleaned up.
and other customer information by breaking The duplicate cookie on the Web server
into a Web server. Those who are referred remains after a user deletes a cookie from
to as “script kiddies” may merely aim to their computer. Even if the user is cautious
damage a website by adding text or images. to accept only temporary cookies or none at
 INTERNETWORKS 173
all, the server may be able to “identify” them enable users to browse the internet. When
when they return to the website later on using the website, a window displaying
thanks to a variety of system and software the requested pages opens. The website’s
data that browsers frequently provide. IP address is the only one visible on the
Internet. However, it is important to note
Because of these worries, a lot of astute that the website itself will log and monitor
internet users would rather browse in users’ IP addresses.
secret. Here, a proxy server is positioned
in between the user’s PC and the Internet.
Every browser request is routed through 4.3.8. Difference between Public
the proxy and then sent to the Internet. and Private IP Addresses
Instead of logging the proxy’s IP address,
web servers return pages to it. After All computers connected to a network,
receiving the page, the proxy serves as a whether public (such as the World Wide
go-between by forwarding it to the surfer. Web) or private (such as those in a
Although the proxy server will track and workplace), are uniquely identified by their
record IP addresses, web servers do not Internet Protocol addresses, or IP addresses.
have a record of the surfer’s IP address. Computers can locate and communicate
It’s critical to ascertain whether a proxy with each other thanks to the IP addressing
service is actually anonymous before using scheme. Each computer connecting to the
it. Some proxy servers thwart the intent by Internet is assigned a public IP address,
sending the requester’s IP address through while computers within the same local
their headers. It is only anonymous proxy area network (LAN) are recognized by their
servers that conceal this data. It is advised private addresses. This system operates
to personally verify through the proxy tools similarly to how phone numbers work in
that are available because some proxies an office environment. A company typically
make false claims about being anonymous. has a primary business number (the public
Since the netizen’s home country will not IP address), and each employee may have
have jurisdiction over the proxy’s logs, using a distinct extension that enables calls to
international proxies can further enhance be directed to the correct individual (the
anonymity. Nevertheless, rather than private IP address).
shielding unlawful activity, proxy services An IP address, a set of binary numbers,
are meant to support the preservation of is used to identify a computer or other device
freedom and privacy for legitimate activity. on a network and provide information about
To prevent records from accumulating it. In the more commonly used addressing
on a single proxy server, users can maintain scheme, IP Version 4 (IPv4), these numbers
a list of proxies and quickly switch between are typically expressed as four numbers
them using the FoxyProxy add-on plug- separated by dots. A new addressing
in for the FirefoxTM browser. Additionally, scheme, IPv6, was developed in the 1990s
proxies can be linked by chaining two or due to the limited and depleting number of
three proxy servers together, but this slows addresses available in the IPv4 format. In
down browsing. Moreover, requests are lost this format, eight groups of four letters and
if a proxy server is unavailable. For proxy numbers are used to represent IP addresses;
browsing, there are also various shareware groups with a zero value may be omitted.
applications available. Through an on-site Groups are separated by colons. In IPv6,
interface, web-based anonymous services private addresses are known as “local-use.”
174 COMPUTER NETWORKS

• Public Addresses: The IP addresses that are public enable

any two computers to recognize one another. A person’s
computer is typically assigned an address from a pool
that has been reserved for her Internet Service Provider
(ISP) to use for its clients when they connect to the
Internet. The IP address of the server that hosts the
website is generated when she types in a domain name,
such as wisegeek.com. To determine where to send the
requested site page, the server uses the computer’s
public IP address.
Keyword • Private Addresses: A private network can be set up by
connecting multiple computers or devices wirelessly or
Internet service
provider (ISP) is through cables. To enable file sharing and resource sharing
an organization that across the network, each device is assigned a unique IP
provides services address. Different private networks may use the same
for accessing, using,
or participating in addresses, as long as they are unique within their own
the Internet. network; this is because computers in different networks
do not communicate directly with each other. Using private
IP addresses as identifiers, a network router transfers data
between the connected computers. A modem is usually
used to link the private network or one of its computers to
the Internet. The Internet Service Provider (ISP) assigns a
public IP address to the router or firewall in the network;
this single public IP address represents the entire network
on the Internet. The router acts as a gatekeeper, directing
requests from specific computer users to the Internet
through a device called a Network Address Translator
(NAT). The router decides how to send returned data to
the public IP address.
• Static and Dynamic: A public IP address may be either
dynamic or static. Websites and other online services
are typically hosted on static IP addresses, which remain
constant. Because they can facilitate communication, some
Voice over IP (VOIP) users and gamers also favor static
IPs. Every time a certain user connects to the Internet,
a dynamic IP address is selected from a pool of available
addresses. Since it’s the default address type assigned to
users of an ISP when they use its services, the majority
of computers have dynamic public IP addresses.
• Finding a Computer’s IP Address: Users can view the public
IP address assigned to their computer or, occasionally,
the IP addresses of other users, using a variety of online
programs. Via the configuration dialogs on the network
router, private IP addresses can also be seen. Before trying
 INTERNETWORKS 175
to access or modify the data in a network router, inexperienced computer users
might want to speak with a network administrator or another expert.

Did you know?

Classful network design served its purpose in the startup stage of the Internet, but it lacked scalability in the face of
the rapid expansion of networking in the 1990s. The class system of the address space was replaced with Classless
Inter-Domain Routing (CIDR) in 1993. CIDR is based on variable-length subnet masking (VLSM) to allow allocation
and routing based on arbitrary-length prefixes.

4.3.8.1. Private Ranges

Within the range of publicly available IP addresses there are specific, excluded ranges
withheld for private network use. In IPv4, the private IP ranges are as follows:
• 10.0.0.0 ... 10.255.255.255
• 172.16.0.0 ... 172.31.255.255
• 192.168.0.0 ... 192.168.255.255
In IPv6, site-local addresses—the equivalent of IPv4 private addresses—begin with
FE followed by C, D, E, or F. Another type, called link-local, does not have a comparison
in IPv4 and is only used for special purposes on physical networks. These also begin
with FE, followed by 8, 9, A, or B.

4.3.8.2. Who Coordinates IP Addresses?

The global assignment of IP addresses and other associated protocols is governed by
the Internet Assigned Numbers Authority (IANA). The Internet Corporation for Assigned
Names and Numbers (ICANN) is currently in charge of IANA, which was previously an
independent organization. The organization in charge of allocating and managing Internet
domain names, such as those ending in .com and .org, is ICANN. The challenging task
of connecting computers to each other so they can efficiently exchange information is
facilitated by IP addresses, domain names, and other identifiers.

Did you get it?

1. Can you explain the concept of website IP addresses and their role in hosting web content
on the internet?
2. What tools and techniques are used to track and record IP addresses, including network
monitoring software and log files?
176 COMPUTER NETWORKS

ROLE MODEL
Paul Baran: He was one of the two independent inventors of packet
switching, which is today the dominant basis for data communications
in computer networks worldwide, and went on to start several
companies and develop other technologies that are an essential
part of modern digital communication.

BIOGRAPHY
Paul Baran (born April 29, 1926, Grodno, Pol. [now Hrodna, Bela.]—
died March 26, 2011, Palo Alto, Calif., U.S.) was an American electrical
engineer and the inventor of the distributed network. He worked
contemporaneously with British computer scientist Donald Davies on
data packet switching across distributed networks. These inventions
laid the foundation for the Internet.
In 1928, Baran’s family moved to Philadelphia. He studied
electrical engineering at Drexel University in Philadelphia (B.S.,
1949) and at the University of California, Los Angeles (M.S., 1959).
In 1959, he joined the RAND Corporation as a researcher, a think
tank that provided analyses of various issues affecting public policy
and national defense. At RAND, Baran focused on developing a
method for U.S. authorities to communicate in case their centralized
switching facilities were destroyed by a nuclear attack. Influenced
by the principle that the human brain can recover lost functions
by bypassing a dysfunctional area, Baran conceived a “distributed”
network using digital technology. This network would have no
centralized switches or dedicated transmission lines and would
continue to operate even if several of its switching nodes were
disabled.
For transporting messages across this system, Baran conceived
of the idea of breaking large messages or units of computer data
into “message blocks”—separate pieces of data that would be sent
independently to the target destination, where they would be rejoined
into the original message. By foregoing dedicated communication
lines in favor of using any number of available circuits, Baran’s
system increased transmission capacity (bandwidth) and created a
flexible, reliable, and robust communications network. Baran’s work
on message blocks appeared in a series of RAND studies published
between 1960 and 1962. At about the same time, Davies in the
United Kingdom invented a similar system employing what Davies
 INTERNETWORKS 177

called “packets,” and packet switching, as this process came to be called, formed the
basis for communication across modern networks. With digital computers as network
nodes, Baran used a “rapid store and forward” design for packet switching, allowing
for essentially real-time data transmission. In the 1970s, Baran became an informal
consultant to ARPANET, a high-speed computer network created by the Defense Advanced
Research Projects Agency (DARPA) to connect research institutes and laboratories
supported by the Department of Defense across the United States. Baran’s inventions
provided the technical foundation for the eventual development at ARPANET of the
Transmission Control Protocol/Internet Protocol (TCP/IP), a communications protocol
that allowed a number of different networks designed by different vendors to form a
“network of networks.” ARPANET, based on Baran’s packet switching, thus became
the predecessor of the Internet.
Baran left RAND in 1968 and afterward was involved with developing discrete
multitone technology (a crucial component of digital subscriber lines) and with contributing
to developments in spread spectrum transmission (an essential component of wireless
communication). Baran also founded Metricom, a wireless Internet service company, in
1986; Com21, a supplier of cable modem systems, in 1992; and GoBackTV, a company
specializing in infrastructure equipment for television operators, in 2003.
178 COMPUTER NETWORKS

SUMMARY
• The term “internetworking” is a compound of the words “inter” and “networking,”
which denotes a connection between completely unrelated nodes or segments.
The intermediary devices, which are similar to routers or gateways, establish
this connection.
• The process of connecting multiple networks via devices such as routers and
gateways is known as internetworking. The various networks are owned by
different organizations with vastly different network technologies.
• Each physical network connection of a network device is uniquely identified by
its data-link layer address. Physical or hardware addresses are commonly used
to refer to data-link addresses.
• Network addresses, also known as virtual or logical addresses, can sometimes
be found within an address space that can be graded. A network address and
a device have a logical, non-fixed relationship that typically depends on either
the physical properties of the network or on arbitrary groupings.
• An internetwork is a group of broadcast and packet-switching networks that
function as a single, large network thanks to the use of routers, switches, and
bridges, which are intermediate networking devices.
• In order to enable efficient communication between an application program and
another application program on a network, the application layer offers services
to the program. It is not appropriate to think of the application layer as an
application in the sense that most people do.
• A physical cable marks the beginning of a network, while a computer program
wraps it up. During a regular data transfer process, information undergoes
various transformations at different levels..
• The Internet layer organizes data into IP datagrams for transmission, containing
source and destination addresses. It also handles routing of IP datagrams.

MULTIPLE CHOICE QUESTIONS

1. Network layer at source is responsible for creating a packet from data coming from
another
a. Station
b. Link
c. Node
d. Protocol
2. Internet has chosen datagram approach to switching in the
a. Data Link Layer
b. Network Layer
c. Physical Layer
 INTERNETWORKS 179
d. Presentation Layer
3. In Internetwork, network layer is responsible for
a. Host to User Link
b. Host to Host Delivery
c. User to Host IP
d. Host to Server communication
4. Datagram network uses universal addresses defined in network layer to route
packets from source to the
a. Same source
b. Layers
c. Destination
d. Application
5. Physical and data link layers of a network operate
a. Locally
b. Seperately
c. Unjointly
d. Independently

REVIEW QUESTIONS
1. Identify the challenges to internetworking.
2. Give an overview on internetwork addressing.
3. What are the importance of application layer and transport layer in internetworking?
4. How to IP addresses work?
5. Define the TCP/IP network architecture.

Answer to Multiple Choice Questions

1. (d) 2. (b) 3. (b) 4. (c) 5. (b)

REFERENCES
1. Advances in Internetworking, Data & Web Technologies, (2017). The 5th International
Conference on Emerging Internetworking, Data & Web Technologies, EIDWT-2017.
Wuhan, China.
2. Comer, D., (2013). Internetworking with TCP/IP (Vol. 1, 6th edn.). Saddle River, NJ:
Pearson.
3. Coulouris, G., Dollimore, J., Kindberg, T., & Blair, G., (2012). Distributed Systems:
Concepts and Design. Addison-Wesley.
180 COMPUTER NETWORKS

4. Kozierok, C., (2005). The TCP IP Guide: A Comprehensive, Illustrated Internet

Protocols Reference. San Francisco: No Starch Press.
5. Medhi, D., (2007). Network Routing: Algorithms, Protocols, and Architectures. San
Francisco: Morgan Kaufmann Publishers.
6. Osterlow, H., (2012). IP Routing Primer Plus. Indianapolis: Sams Publishing.
7. Peterson, L. L., & Davie, B. S., (2012). Computer Networks: A Systems Approach.
Elsevier, Inc.
CHAPTER
Networking
5 for Clusters

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Explain the principles of internetworking

• Define the internetwork architecture
• Discuss on IP addressing and architecture
182 COMPUTER NETWORKS

INTRODUCTORY EXAMPLE

CLUSTER NETWORKS WITH INSTANCE POOLS

Cluster networks with instance pools are a modern approach to efficiently managing
resources in cloud computing environments. This paradigm combines the concepts of
cluster networking and instance pooling to enhance scalability, reliability, and cost-
effectiveness. In this configuration, a cluster network refers to a group of interconnected
servers or virtual machines (instances) that collaborate to perform various tasks or
services. These instances are organized into pools, which are sets of identical or similar
instances configured to handle specific workloads or applications.
One key aspect of cluster networks with instance pools is their dynamic nature.
Instances within a pool can be added or removed based on demand, allowing the system
to automatically scale resources up or down in response to changing requirements. This
elasticity ensures optimal resource utilization and performance efficiency, as resources
are provisioned and de-provisioned as needed. Additionally, instance pools can improve
fault tolerance and reliability by distributing workloads across multiple instances. If one
instance fails or becomes unavailable, the workload can be seamlessly transferred to
another instance within the pool, minimizing downtime and service interruptions.
Furthermore, cluster networks with instance pools offer cost benefits by optimizing
resource usage. Instead of provisioning separate instances for each workload, organizations
can leverage shared pools of instances, reducing the total number of resources required.
This consolidation leads to lower infrastructure costs and better overall resource utilization.
Additionally, cloud providers often offer pricing models based on resource usage, allowing
organizations to pay only for the resources they consume.
INTRODUCTION
A cluster network is an ultra-low latency, high-bandwidth network that connects a pool
of GPU-optimized instances or high-performance computing (HPC) instances. Every node
in the cluster is a simple machine made of metal that is situated near other nodes.
Similar to on-premises HPC clusters, a remote direct memory access (RDMA) network
between nodes offers latency as low as single-digit microseconds.
Cluster networks are designed for highly demanding parallel computing workloads.
For example:
• Computational fluid dynamics simulations for automotive or aerospace modeling
• Financial modeling and risk analysis
• Biomedical simulations
• Trajectory analysis and design for space exploration
• Artificial intelligence and big data workloads
The instance pools feature serves as the foundation for cluster networks. Although
you can monitor, add tags, and resize the underlying instance pool, the cluster network
manages the majority of instance pool operations.

NETWORKING FOR CLUSTERS 183

5.1. CLUSTER-BASED
NETWORKS
→Learning Objectives
• Identify the components and characteristics of cluster-based network architectures
• Implement and configure routing protocols to facilitate efficient data transmission
and communication within cluster-based networks

Large dynamic networks can be controlled more effectively by using resources thanks
to cluster-based network control structures. They make excellent candidates for ad hoc
networks as a result. Cluster-based architectures come in various varieties. In particular,
the link-cluster architecture is made for managing transmission in ad hoc networks.
The clustering techniques used for backbone formation are virtual subnet architecture
and Near-Term Digital Radio Network (NTDR) architecture. The preferred method for
increasing routing efficiency in cluster-based networks is hierarchical routing, whether
it is strict or quasi-hierarchical.
An entity with dynamically changing states in an unpredictable way is a communication
network. A network must be able to modify its behavior to account for variations in
its inherent characteristics, such as connectivity, capacity, and offered load, in order to
achieve performance targets set for user traffic. The following goals are in competition
with each other for control functions that govern the network’s performance:
• Quick and accurate response when adapting the network’s behavior to the
current network state.
• Minimal use of network resources during and after the adaptation process.
Self-organizing controlled structures work better for networks with fluctuating
connectivity (e.g., mobile networks) or those that are large with numerous configuration
parameters. Nodes construct and maintain these structures, leading to increased network
availability, faster response to state changes, and reduced risk of configuration errors.
Depending on the control functions to be carried out, the size of the network, and the
anticipated frequency and magnitude of changes in network state, the proper network
control structure and algorithms for using this type of structure are selected.
Applicability of these structures and algorithms to ad hoc networks is also investigated,
particularly on routing functions. A cluster-based control network improves efficiency
of resource use by creating contexts for:

184 COMPUTER NETWORKS

 NETWORKING FOR CLUSTERS 185
• Managing wireless transmission among multiple nodes to reduce channel
contention.
• Forming routing backbones to reduce network diameter.
• Abstracting network state information to reduce its quantity and variability.

5.1.1. Cluster Based Network Architectures

The cluster based control structures used in ad hoc networks to achieve specific purposes
are described below.

5.1.1.1. Link-Cluster Architecture

Broadcast environments with multiple access use this kind of architecture. Different
node clusters are arranged so that transmissions are controlled without contention,
which lowers interference. All network nodes in this architecture autonomously group
together into interconnected clusters. As seen in Figure 5.1, each cluster consists of a
cluster head, one or more gateways, and zero or more regular nodes. The cluster head
distributes resources among the clusters and plans transmissions. Adjacent clusters are
linked by gateways. A gateway can form an indirect connection between two clusters
by joining one and creating a link to a member of the other, or it can form a direct
connection between two clusters as members of both. As a result, the architecture may
have clusters that overlap or are disjoint.

Figure 5.1. The Link-Clustered Architecture (Source: Springer, https://media.

springernature.com/m685/springer-static/image/art%3A10.1007%2Fs12083-021-01175-6/
MediaObjects/12083_2021_1175_Fig1_HTML.png).

Nodes establish a link-clustered architecture over a physical network by the following:

1. Discovering neighbors with bi-directional connectivity by broadcasting a list of
those neighbors that they can hear and by receiving broadcasts from neighbors.
186 COMPUTER NETWORKS

2. Electing cluster heads and forming clusters.

3. Agreeing on gateways between clusters.

5.1.1.2. Cluster Heads

For the purpose of electing cluster heads, at least two algorithms—
identifier-based clustering and connectivity-based clustering—have
been proposed. Both algorithms can be implemented centrally or
distributedly. In the centralized version, the cluster head for the
cluster that consists of that node and its one-hop neighbors is
determined by looking at the node with the greatest number of
neighbors (connectivity-based) or the lowest number of neighbors
(identifier-based). A node that has the lowest or highest numbered
identifier in its neighborhood elects itself when using the distributed
version of identifier-based clustering. If a node is the most highly
connected among all of its neighbors who are not chosen to be
cluster heads, it becomes a cluster head in the distributed version of
connectivity-based clustering. The literature also describes variants
of clustering algorithms that, for the two methods mentioned above,
always form disjoint clusters. In those situations, cluster heads serve
as coordinators of intra-cluster transmission only, not as facilitators
of cluster formation.

5.1.1.3. Node Mobility

To ensure accurate transmission scheduling, cluster membership
must be updated in line with node movements within the network.
Whenever a node joins or leaves a cluster, the clustering algorithms
as described by Baker, Ephremidis, Wieselthier, Gerla, and Tsai
recalculate the cluster membership, cluster head, and gateway status.
In the cluster recompilation approach, identifier-based clustering
proves to be more stable than connectivity-based clustering. This
Keyword is attributed to the potential for more frequent changes in a node’s
Connectivity- cluster head status when utilizing connectivity-based clustering. The
based clustering algorithm that minimizes cluster changes reduces the number of
is based on the cluster head status alterations caused by node movements. Changes
core idea of objects
being more related in cluster head status only occur when two cluster heads come
to nearby objects within range of each other, leading one of them to relinquish their
than to objects cluster head status, or when a regular node moves out of range of
farther away.
any other node and assumes the role of cluster head for its own
cluster. Lin and Gerla also propose cluster maintenance strategies to
minimize the impact of node movements on the existing cluster set.

5.1.1.4. Routing
A naturally occurring routing backbone made up of cluster heads,
 NETWORKING FOR CLUSTERS 187
gateways, and the links connecting them is provided by the link-clustered architecture.
Nevertheless, cluster heads could become clogged as hubs for traffic, and each one could
end up being a communication breakdown point for the cluster as a whole. For this
reason, routing algorithms do not use link-clustered architecture as their routing control
architecture. Rather, every node creates and chooses routes, distributes and gathers
routing data. In a link-clustered architecture, clusters serve as a routing backbone and
are primarily used to define regions for transmission management.

5.1.1.5. Near-Term Digital Radio (NTDR)

Large tactical networks are where Network Near-Term Digital Radio (NTDR) networking
is meant to be applied and deployed.. This is one of the backbone formation clustering
techniques. As seen in Figure 5.2, NTDR generates a collection of clusters, each with a
cluster head, which are connected to form a routing backbone. It is similar to link-clustered
architecture in that it consists of a single level of clusters with nodes within one hop of
a cluster head. NTDR architecture is not the same as link-clustered architecture in the
following ways, though. Cluster heads serve as intermediaries for direct communication
between clusters. All communication within the cluster must pass through the cluster
head, with the exception of nodes that are adjacent to each other and only require a
single hop.

Figure 5.2. The NTDR Network Architecture (Source: Researchgate, https://www.researchgate.net/

profile/Symeon-Papadopoulos/publication/284104416/figure/fig4/AS:667840692617225@153623703
7913/Hub-specificity-calculation-Blue-lines-indicate-visual-edges-with-weight-exceeding-th-d.png).

Any node has the ability to take over as cluster head quickly in the event that frequent
node movements or cluster head outages necessitate changes in node interconnectivity.
By periodically broadcasting and receiving beacons, each node maintains bi-directional
connectivity with its neighbors.

5.1.1.6. Cluster Heads

If an NTDR node receives beacons advertising two different partition identifiers or
does not receive any beacons from any cluster heads, it elects itself as a cluster head.
The NTDR algorithm uses the following mechanisms to limit the number of nodes
simultaneously attempting to become cluster heads:
188 COMPUTER NETWORKS

• Each node that detects one of the backbone with data and uses that data to
conditions for becoming a cluster compute routes to other network nodes. A
head waits a short random time “resistance” metric, which quantifies the
interval and tests the conditions likelihood of interference that upcoming
again. If the condition persists transmissions over the links may experience,
following the waiting period, the is included in the link state. Cluster heads
node assumes the role of a cluster use Dijkstra’s shortest path-first (SPF)
head. algorithm to calculate least resistance
• Each new cluster head immediately routes to destinations.
issues beacons in quick succession
proclaiming its status. 5.1.1.9. Virtual Subnet Architecture

5.1.1.7. Cluster Affiliation In a multihop mobile wireless network, the

virtual subnet architecture uses a number of
A node seeking cluster affiliation prefers disjoint routing backbones to provide fault-
clusters in which tolerant connectivity and load balancing.
The network is first split into a number of
• The node and the cluster head
physical subnets of disjoint clusters based
belong to the same organization.
on the node locality. Virtual subnets, which
• The signal from the cluster head span all physical subnets, are created by
is transmitted at low power but grouping together members of various
received at high strength. physical subnets.
• The resulting cluster size is
relatively small.
• A cluster member remains affiliated
with its chosen cluster head until
one of the following occurs:
• The cluster head relinquishes its
role.
• The cluster head’s beacons no
longer list the member.
• The received signal strength from
the cluster head is unacceptably
low.
Figure 5.3. The virtual Subnet Architecture
5.1.1.8. Routing (Source: S e m a n t i c s c h o l a r, https://
d3i71xaburhd42.cloudfront.net/df535cc40f672e
The cluster heads in the NTDR network are
2f9003a73de4ac8facb58f9e5a/3-Figure3-1.png).
in charge of keeping the backbone stable, so
they keep an eye on it and share information To minimize interference, distinct
about any changes with one another. To frequencies are assigned for communication
communicate with other cluster heads, between adjacent physical subnets and
each cluster head generates link state all virtual subnets. In the network, the
information about its links and membership maximum number of virtual subnets, Q, and
information about its cluster. It floods the the maximum number of physical subnets, P,
 NETWORKING FOR CLUSTERS 189
are predefined. As illustrated in Figure 5.3, address will be sent back to the source.
every node belongs to exactly one physical Otherwise, the source will distribute the
subnet and zero or more virtual subnets. address query within its virtual subnet.
If a node communicates regularly with the If the destination is currently linked to a
members of multiple virtual subnets, it may physical subnet indicated by some, the
be considered a member of those subnets. destination’s address will be returned.
The prefix of a node’s address, which
represents the physical subnet, and the 5.1.1.11. Routing
suffix of its address, which represents
When there are multiple routes available
the virtual subnet, allow it to be uniquely
within a virtual subnet architecture, a
identified as a node within this cluster-
source can more quickly and effectively find
based control structure. A node will have
a different route to their destination and,
multiple addresses and address suffixes if it
as a result, maintain their current session
is a member of multiple virtual subnets. An
even when nodes move and the network’s
ad hoc network can be organized and used
connectivity changes. Two approaches are
with multiple overlaid routing backbones
recommended for multiple forwarding. First,
using the virtual subnet architecture.
there is direct routing, where packets are
Therefore, before it could be realized as
forwarded to the destination based only
an actual ad hoc network, a full set of
on source and destination addresses from
networking algorithms compatible with
the source. Packets are forwarded through
this architecture, such as subnet clustering,
a node if it is discovered in a manner
frequency assignment, distribution of
that allows it to be a part of both the
address and routing information, computing
destination’s virtual subnet and the source’s
of routes, and packet forwarding, would
physical subnet, as indicated in Figure
need to be specified.
5.4(a). As seen in Figure 5.4(b), packets
are routed via a node that is a member of
5.1.1.10. Node Mobility both the destination’s physical subnet and
the source’s virtual subnet if the destination
A node might have to join a new virtual node is a guest of a physical subnet.
subnet and another physical subnet when
it moves to remain connected to the However, in the presence of highly
network. If all nearby physical subnets mobile nodes, this straightforward routing
have reached their virtual subnet limits, might not be feasible. The second technique
the node can connect as a “guest” to is long-path routing, which helps the
one of them. If there’s space available, network support divided subnets and
the node can then become a member. It distribute traffic load evenly among nodes
will then inform all members of its new by randomly distributing the routes chosen
physical and virtual subnets about its new over the space of feasible routes. One of
address. The source node uses a unique the virtual subnets contained in its physical
identifier to distribute an address query subnet is randomly selected by the source.
within its physical subnet before trying to Packets will be routed to a node that is a
communicate with the destination node. If member of both the selected virtual subnet
the destination is associated with a virtual and the physical subnet of the source if
subnet represented by a member of the it is not a member of that virtual subnet.
source’s physical subnet, the destination’s As seen in Figure 5.5(a), routing is then
190 COMPUTER NETWORKS

carried out to a member node of the destination’s physical subnet

as well as the virtual subnet of choice. As demonstrated in Figure
5.5(b), if the source is a part of the selected virtual subnet, it will
arbitrarily choose a different physical subnet from its own that is
Remember represented within the virtual subnet and route packets through
If every subnet it to the destination.
is connected,
long-path routing
generates at
most Q+P-1
distinct routes
from a source
to a destination.
The number
of possible
routes increases
when there
are partitioned
subnets because
each intermediate
node along the
route might
have to choose
at random the
next hop subnet
to route around
such partitions.
Figure 5.4. Direct Routing (Source: Semanticscholar, https://d3i71xaburhd42.
cloudfront.net/df535cc40f672e2f9003a73de4ac8facb58f9e5a/4-Figure4-1.
png).

Figure 5.5. Long-Path Routing (Source: Semanticscholar, https://

d3i71xaburhd42.cloudfront.net/df535cc40f672e2f9003a73de4ac8facb58f9e
5a/5-Figure5-1.png).
 NETWORKING FOR CLUSTERS 191
Consequently, a route’s length may
grow significantly. Nonetheless, nodes
can shorten routes by providing additional
details about their interconnectivity and
utilizing those details to choose a shorter
path. As a result, nodes may choose a
routing strategy that combines the first
two mentioned above.

5.1.2. Routing in Cluster-Based

Networks
The inherently dynamic nature of ad hoc
networks poses challenges in the design
of effective yet efficient routing functions.
Routing functions are the focus of this
section.
Figure 5.6. The Nested Cluster Architecture
5.1.2.1. Hierarchical Routing (Source: S e m a n t i c s c h o l a r, https://
d3i71xaburhd42.cloudfront.net/df535cc40f672e
A hierarchical cluster-based control structure 2f9003a73de4ac8facb58f9e5a/5-Figure6-1.png).
is a network consisting of N nodes organized
One can categorize the hierarchical
into an m-level hierarchy of nested clusters
routing scheme as strictly hierarchical or
of nodes, as shown in Figure 5.6, such that
quasi-hierarchical. The following example
all level-i clusters are disjoint for 0≤i<m.
clarifies how they differ from one another.
Every node is a cluster at level 0. They Assume for the moment that the lowest-
are organized into level-1 clusters, which level cluster, ck, contains a session’s source
are then organized into level-2 clusters, and destination. A packet is routed directly
and so on, until all level-(m-1) clusters are from the source, s0, to the boundary of dk-
finally organized into a single level-m cluster 1, the level-(k-1) cluster of the destination,
that is made up of every node. This kind of d0, when using quasi-hierarchical routing.
nested cluster structure uses a hierarchical After that, the packet is routed straight to
addressing scheme. The concatenation of the dk-2 boundary, the level-(k-2) cluster of
the labels of the level-(m-1) through level-0 d0, and so on, until it is ultimately routed
clusters that contain the node is how the straight from the d1 boundary to d0. A
address is expressed. For instance, the packet is routed indirectly from s0 to the
address of node z in the clustering hierarchy edge of sk-1, s0’s level-(k-1) cluster, using
shown in Figure 5.6 is x.y.z, with u acting as strict-hierarchical routing. Initially, the
the implicit prefix. The disjointness criterion packet is sent straight from s0 to s1’s edge.
is not a prerequisite for routing schemes to It is then routed via s2’s level-1 clusters
function properly in the presence of non- until it reaches the s2 boundary, and so
ancestral clusters. on, until it reaches the sk-1 boundary. The
packet is routed through level-(k-1) clusters
in ck to reach the boundary of dk-1 after it
reaches the boundary of sk-1. The packet
192 COMPUTER NETWORKS

is then routed through dk-1’s level-(k-2) to other nodes in the cluster. Next, using
clusters to get to dk-2’s border, and so on, Dijkstra’s SPF algorithm, each node creates
until it is ultimately routed to di. a minimum-cost route to every other node
in the cluster. Every gate node on a cluster
boundary computes every other gate for
the cluster using intra-cluster link-state
information. Next, a gate builds link-state
information in the form of costs over virtual
links to “neighboring” gates, both in the
gate’s immediate vicinity and in neighboring
clusters that are directly connected to it.
This link-state data is flooded by the gate
to every other gate in every level-1 cluster.
Next, every gate within the network creates
the least expensive paths to every other
gate, and consequently, to every other
cluster within the network. Additionally,
every gate floods all other clusters with
Figure 5.7. Quasi-hierarchical Routing its minimum costs through its intra-cluster
versus Stricthierarchical Routing (Source: nodes. A node can ascertain the n by fusing
Semanticscholar, https://d3i71xaburhd42. this intra-cluster link-state information with
cloudfront.net/df535cc40f672e2f9003a73de4ac8 the inter-cluster cost information.
facb58f9e5a/6-Figure7-1.png).

5.1.2.2. Quasi-Hierarchical Routing

The quasi-hierarchical category includes the
majority of hierarchical cluster-based routing
systems. Moreover, the distance vector
routing technique is the basis for nearly
all quasi-hierarchical routing schemes. In
this case, the objective is to determine the
next hop on the minimum-cost route from
a node to each level-i cluster within the
node’s level-(i+1) cluster, for 0≤i<m. The
information about routing inside the cluster
is not distributed outside the cluster.
Link-state routing can also serve
as the foundation for quasi-hierarchical Figure 5.8. The Link-State Routing Architecture
routing. Nodes create, share, and utilize with Gates (Source: Citeseerx, https://
hierarchical routing information in the d3i71xaburhd42.cloudfront.net/df535cc40f672e
following ways when using this scheme. 2f9003a73de4ac8facb58f9e5a/6-Figure8-1.png).
Every node in a level-1 cluster floods its
link-state information, which is expressed Large tactical packet radio networks
as the link cost to its neighboring nodes, can use hybrid routing schemes, such
as distance-vector and link-state quasi-
 NETWORKING FOR CLUSTERS 193
hierarchical routing schemes. This hybrid plan was proposed for the
SURAN (Survivable Adaptive Networks) initiative. The next hop on
the least-cost route to each destination is ultimately determined by
each node using distance-vector routing in this scheme; however,
for destinations outside of its cluster, the distance-vector routing
Keyword
Survivable
information is partially derived from link-state data between clusters Adaptive
and superclusters. Certain nodes are used as routing focal points Networks
in quasi-hierarchical routing schemes like landmark routing and (SURAN) refer to
networks designed
regional node routing. When a packet is being forwarded from to withstand and
source to destination, the focal nodes assist other nodes in making adapt to various
decisions about it. Every level-k cluster, also known as a “k-region,” types of disruptions
or attacks, ensuring
has one or more k-regional nodes, with the exception of the top- continuity of critical
level m-region, when using regional node routing. Figure 5.9 below communication and
shows the addressing scheme and node affiliations to regions. services.

Figure 5.9. The Regional Node Routing Architecture (Source: Citeseerx,

https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=df535cc
40f672e2f9003a73de4ac8facb58f9e5a).

With landmark routing, each level-i landmark, x, defines a level-i

cluster consisting of all nodes that are within the radius of ri(x)
node hops from x, where 0 < ri(x) ≤ ri and ri is the maximum radius
194 COMPUTER NETWORKS

permitted for the level-i cluster. Each level-m level-i cluster, c, must be siblings within
landmark has a radius that is at least as x’s level-(i+1) cluster. If c’ and c are non-
large as the diameter of the network, and neighboring siblings, the next hop to c’ is
each node is a level-0 landmark. For 0 < i a level-i cluster that is a neighbor of c. If
≤ m, more than one level-i landmark is also c’ and c are neighboring siblings, the next
a level-(i+1) landmark, and for each such hop to c’ is a level-j cluster, where 0≤j<1,
landmark, y, ri+1(y) > ri(y). that lies on the boundary of c and contains
at least one node, y, with a direct link to c’.
With quasi-hierarchical routing, a node can
5.1.2.3. Strict-Hierarchical Routing determine the next hop to any destination
Because it is resilient to changes in the using a single forwarding table entry,
network state, strict-hierarchical routing whereas with strict-hierarchical routing,
is a preferred method in mobile wireless a node may have to consult up to 2m-1
networks. Comparing the robustness to forwarding table entries to determine the
quasi-statistical routing, higher route costs next-hop node to a destination. In practice,
and packet forwarding overhead result m remains small.
from the robustness. Using either the The expenses of the routes that the
distance-vector or the link-state approach, forwarding entries are derived from are
a representative of a level-i cluster, c, the subject of the second distinction.
gathers its routing data and distributes it Since the cost to reach the level-i cluster,
to representatives of other level-i clusters c’, from x is calculated as the sum of the
within its level-(i+1) cluster. Within its level- individual link costs, it is always the true
(i+1) cluster, c’s representative calculates minimum cost when using quasi-hierarchical
the next-hop clusters and minimum-cost routing. The total of each cluster cost for
routes from c to any other level-i clusters each level-i in the route from c to c’ is
using routing information similarly obtained the cost to reach c’ from x when using
from other clusters. The representative of c strict-hierarchical routing. As a result, the
then sends the cluster boundary information, latter routing is probably going to produce
route cost, and next-hop cluster information more expensive routes than the former one.
to every node inside c. For the SURAN packet radio network, the
There are two key differences strict-hierarchical routing scheme below
between strict-hierarchical routing and was selected due to its ability to maintain
quasi-hierarchical routing with respect to routes in the face of dynamically changing
forwarding information. The first difference nodes and links. To find the lowest cost
concerns the number of forwarding table routes and next-hop nodes to every node in
entries a node must consult to determine its cluster and to every neighboring cluster
the next-hop node to a destination. With within its supercluster, each node employs
quasi-hierarchical routing, the next hop to distance vector routing. The cluster head,
level-i cluster c’ stored in the forwarding also known as the super-cluster head, is a
table is always a node which is a neighbor node within each cluster or supercluster,
of node x. With strict-hierarchical routing, as depicted in Figure 5.10. This node is
the next hop to c’ stored in the forwarding in charge of creating and disseminating
table is always a cluster at level j≤ i, as routing information about the cluster to all
explained below. For an entry for c’ to other clusters or superclusters within the
appear in x’s forwarding table, c’ and x’s same supercluster (or network).
 NETWORKING FOR CLUSTERS 195
Did you know?
A 2003 measurement study of Internet routes found that, between pairs of neighboring ISPs, more than 30% of
paths have inflated latency due to hot-potato routing, with 5% of paths being delayed by at least 12 ms. Inflation
due to AS-level path selection, while substantial, was attributed primarily to BGP’s lack of a mechanism to directly
optimize for latency, rather than to selfish routing policies.

The connectivity and cost to each neighboring clusterhead (or super-clusterhead) are
determined by each clusterhead (or super-clusterhead). It then inundates all clusterheads
(or super-clusterheads) within its supercluster (or network) with this link-state data.
The number of node hops to the neighboring clusterhead (or superclusterhead), the
number of cluster hops to the neighboring superclusterhead, and the value 1 to indicate
inter(super)cluster connectivity only are among the suggested costs from one clusterhead
(or super-clusterhead).

Figure 5.10. The SURAN Routing Architecture (Source: Citeseerx, https://citeseerx.ist.psu.edu/

document?repid=rep1&type=pdf&doi=df535cc40f672e2f9003a73de4ac8facb58f9e5a).

In tactical packet radio networks, a hybrid strict-hierarchical routing scheme is

employed in large networks. Regarding forwarding table entries, this routing scheme is
strict-hierarchical; however, with regard to the granularity and dissemination of routing
information, it is quasi-hierarchical. In order to find the least expensive route and the
next hop that corresponds to each node in its level-1 cluster, each node engages in
distance-vector routing. It also adds the discovery of the level-1 cluster’s boundary nodes
and their connectivity to neighboring clusters. The next hop is a neighboring node,
196 COMPUTER NETWORKS

and route cost within a level-1 cluster is expressed in terms of the

number of node hops. Global routing nodes supply the data needed
for routing outside of a node’s level-1 cluster. A global routing node
serves as a cluster representative in each level-1 cluster. The quasi-
hierarchical routing process is carried out by global routing nodes.
The link-based strict-hierarchical routing schemes described
below were designed to provide quality of service routing in large
dynamic networks. Both schemes have the following features:
• Nodes represent and distribute routing information according
to the link-state approach, at all levels of the clustering
hierarchy.
• Nodes select routes with the objective of satisfying individual
users’ service requests given the constraints imposed by
the current state of the network, and they forward packets
as directed by the sources.
Keyword In the context of providing quality of service, the destination-
Virtual gateway based packet forwarding used by hierarchical routing
allows resources
that are outside schemes up to this point is insufficient. Source-directed
of your mesh to forwarding increases the likelihood of successful packet
communicate to delivery by guaranteeing that various nodes make forwarding
resources that are
inside of your mesh. decisions consistent with the route chosen by the source, in
addition to providing quality of service. Furthermore, smaller
forwarding tables are typically the outcome of source-
directed forwarding. A node does not need to keep up a
forwarding table when using explicit-route forwarding. A
node’s forwarding table size in virtual-circuit forwarding
is directly correlated with the number of virtual circuits
it participates in. The number of destinations for which a
node has received routing information directly relates to the
size of the node’s forwarding table when using destination-
based forwarding.
A strict hierarchical routing scheme, or MMWN, is created for
large mobile wireless networks that support multimedia. The MMWN
nested-cluster architecture is shown in Figure 5.11. Every node in
this architecture is a level-0 cluster, but the number of ancestral
clusters varies between two different nodes. As a result, a cluster
may be connected to more than one level in the clustering hierarchy,
and the maximum level connected to the single cluster that contains
every other cluster is how the height of the hierarchy is expressed.
As long as two clusters share the same parent cluster, they are
regarded as siblings even if they appear at different levels of the
clustering hierarchy. If at least one pair of border nodes, one in
each cluster, have a direct link, then sibling clusters may form a
connecting virtual gateway. The border nodes in each of the two
 NETWORKING FOR CLUSTERS 197
clusters along with their connecting links make up a virtual gateway. The endpoints
of a virtual link that spans a cluster are two virtual gateways that are connected to
the cluster. At least one QoS manager is present in every cluster, producing abstracted
link-state data such as connectivity to the cluster, services (e.g., disperses the data
to other clusters, and transfers the data (delay, throughput, loss rate) throughout the
cluster and to nearby clusters. A cluster’s QoS manager broadcasts link-state updates
to every node in the parent cluster.

Figure 5.11. The MMWN Routing Architecture (Source: Citeseerx, https://citeseerx.ist.psu.edu/

document?repid=rep1&type=pdf&doi=df535cc40f672e2f9003a73de4ac8facb58f9e5a).

A node makes an effort to create routes that are compatible with the users’ service
requests and the network’s existing capabilities using this link-state information. There
exists an opportunity to greatly minimize the amount of network state data that
needs to be dispersed, processed, and stored across a network by utilizing hierarchical
cluster-based control structures. It is accomplished by abstracting away the specifics
of a cluster’s connectivity and services, and expressing the cluster’s state in this way.
However, balancing the abstraction’s fidelity and size is necessary to build a workable
abstraction for a cluster whose features vary widely or often. All abstraction techniques,
however, have in common that a cluster’s connectivity and services are expressed in
terms of the cluster’s neighboring clusters, which serve as the cluster’s entry and exit
points.

Did you get it?

1. How does NTDR differ from traditional analog radio technologies, and what are its key
advantages?
2. How can quasi-hierarchical routing be implemented and optimized to improve network
performance and scalability?
198 COMPUTER NETWORKS

5.2. CLUSTERING
PROTOCOLS
→Learning Objectives

• Define clustering protocols in the context of network communication

• Explain the LEACH (Low Energy Adaptive Clustering Hierarchy) clustering
protocol and its significance in wireless sensor networks.
• Explore the concept of clustering and routing protocols tailored for wireless
sensor networks (WSNs).
• Identify the essential components and resources required to establish a cluster
networking environment.

As the object is being tracked, the clustering protocol simplifies the transfer of
information between cluster heads regarding the object’s current state and the uncertainty
of its target position. The topology of the hierarchical, non-overlapping sensor node
clusters is specified by clustering protocols. A robust clustering method is essential for
sensor networks to self-organize. An effective clustering protocol ensures clusters with
nearly equal radii and optimally positioned cluster heads. Route discovery among cluster
heads is sufficient to establish a workable route in a clustered network since each node
is connected to a cluster head. Clustering can reduce the number of transmissions and
facilitate multihop route discovery for large sensor networks compared to a flat, non-
clustered network.

5.2.1. Classification of Clustering Protocols

Techniques for clustering can be decentralized or centralized. Each sensor node must
transmit its unique data, including its location and energy level, to the central base
station in order to use centralized clustering algorithms. A base station determines
the number of clusters, their sizes, and the positions of the cluster heads using a
predetermined algorithm. It then assigns a new duty to each node. Assuming that
sensor networks could have thousands of nodes, it is not feasible, if not impossible, for
a base station to gather data about each node in the network before setting up a route.
For big sensor networks, centralized clustering is therefore not an option. A clustering
algorithm should be able to form clusters without the assistance of the base station
 NETWORKING FOR CLUSTERS 199
and knowledge of node positioning, since a sensor node starts a
clustering process without any knowledge about its location relative
to the corresponding base station. While location-finder devices can
also be used for this purpose, they are frequently too expensive or
add excessive network overhead.
Clusters are formed using decentralized clustering techniques
independently of centralized base stations. Each sensor node has a
probability of p to become a cluster head, and it can be achieved by
an energy-efficient hierarchical clustering algorithm that broadcasts
its candidacy to nodes that are within k hops of the cluster head.
There are limits to a hierarchical structure with an arbitrary number
of levels because wireless sensor nodes have a limited transmission
range. Upper-level cluster heads may become so separated from
one another as the number of hierarchical levels rises that they are
unable to communicate. Two examples of decentralized clustering
protocols are the Decentralized Energy Efficient Cluster Propagation
(DEEP) protocol and the Low-Energy Adaptive Clustering Hierarchy
(LEACH) algorithm.

5.2.2. LEACH Clustering Protocol

The Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol
is a decentralized clustering algorithm that lacks a strategy for
defining the positioning and distribution of cluster heads, making
it an incomplete energy-optimization solution. By periodically re- Remember
clustering and altering the network topology, LEACH is an application- The LEACH
specific protocol architecture that seeks to extend network lifetime. algorithm is
distributed
LEACH is divided into rounds consisting of a clustering phase and
because it can
a steady-state phase for data collection. At the start of each round, be carried out
a sensor node randomly chooses a number between 0 and 1 and locally at each
then compares this number to a calculated threshold called T(n). If node through
T(n) is larger than the chosen number, the node becomes a cluster computation and
communication,
head for the current round. The value T(n) is calculated using the as opposed to
following formula: transmitting the
location and
energy level of
every node to
a single point.
Nevertheless,
where p is the ratio of the total number of cluster heads to the there is
total number of nodes, r is the number of rounds, and G is a set no energy
consumption
of nodes that have not been chosen as cluster heads for the last
optimization, and
1/p rounds. For the first round (r=0), T(n) is equal to p, and nodes cluster heads
have an equal chance to become cluster head. As r gets closer are selected at
to 1/p, T(n) increases, and nodes that have not been selected as random.
cluster head in the last 1/p rounds have more chance to become
200 COMPUTER NETWORKS

cluster head. After 1/p – 1 rounds, T(n) is specified location will be done by hand. For
equal to 1, meaning that all the remaining instance, in the context of gathering data
nodes have been selected as cluster head. during a war, optical distribution (OD) is
Thus, after 1/p rounds, all the nodes have typically achieved through aircraft sprinkling,
had a chance to become a cluster head whereas in the context of gathering data on
once. Since being the cluster head puts a underwater detection, nodes are typically
substantial burden on the sensor nodes, this placed artificially in their surroundings.
ensures that the network has no overloaded Since OD runs solely on batteries, its energy
node that runs out of energy sooner than supply is finite and cannot be increased.
the others. The two nodes were chosen based on the
similarities and differences between their
Following their self-selection, cluster initial energies.
heads begin to promote themselves to other
sensor nodes. A sensor node begins to Routing protocols refer to the means
determine which cluster head it corresponds of data collection and transmission in a
with when it receives advertisements from network. Numerous researchers have put
multiple cluster-head candidates. Every forth different WSN routing protocols in
node pays attention to the signals from the recent years. These routing protocols
advertisements and selects the candidate can be separated into single-hop and
whose signal is received with the highest multi-hop categories based on the data
power. By doing this, every sensor node is transfer between OD and SN. Whereas in
guaranteed to select the nearest candidate the single-hop routing protocol, OD only
to lead the cluster. communicates with SN via single-hop, in
the multi-hop routing protocol, OD and SN
can communicate via single-hop or multi-
5.2.3. Clustering and Routing hop. The hierarchical routing protocols
Protocol for Wireless Sensor originated when the Low-Energy Adaptive
Networks Cluster Hierarchical (LEACH) [1] protocol
suggested the clustering technique for
WSNs are used in a wide range of industries, sensor network development. WSNs divide
including manufacturing automation, the OD distributed into two categories:
industry, data collection during wartime, member nodes (MN) and CH. The primary
and agricultural environment monitoring, duty of MN is to gather data and forward
primarily for the perception and transmission it via a single hop to the relevant CH. The
of environmental data. A WSN is always CH compiles data and sends the message in
made up of one or more sink nodes (SNs), a single or multiple hop fashion to the SN.
which are responsible for uploading and
collecting data, and hundreds of ordinary As WSNs develop, certain issues are
nodes (ODs). The primary responsibility of discovered. To begin with, one of the main
the OD is to perceive environmental data, issues limiting the network’s lifespan is the
which it then transmits to the SN via a imbalance in network load. Within the WSNs,
predetermined method to complete the nodes with the same starting energy are
communication. The OD is inexpensive, but dispersed, and during each data collection
its processing power is constrained. ODs are round, the energy consumption of each node
typically distributed randomly throughout varies. High energy consumption nodes die
the detection space. In certain instances, more quickly, reducing network coverage
the installation of some or all nodes at the or causing SN to become an island, which
 NETWORKING FOR CLUSTERS 201
ultimately results in the failure of WSNs. in 3D space nodes with a distance less than
Second, the issue of “hot point,” which R have only r/R in 2D space. The number
arises from the phenomenon of network of nodes in the area closer to SN decreases
load imbalance, is the situation in which more obviously. When the network coverage
certain nodes in the network experience is broad, the total number of nodes close
significantly higher network loads than to the SN in 2D WSNs will be several times
other nodes because of issues with routing that in 3D WSNs.
protocols or node distribution. These two
issues cannot be avoided by any protocol in The average distance between a
WSNs where nodes are assigned at random. node and the sink increases when the
The only way to lessen it is to try balancing percentage of nodes that are close to the
the network load as much as possible using sink decreases. The LEACH protocol shows
different techniques. a positive correlation between node energy
consumption and inter-node distance. The
When it comes to WSN research, the average distance will eventually rise, which
majority of researchers presume that the will result in higher energy usage and a
application environment is two-dimensional. shorter lifespan for the LEACH protocol. Not
Most WSNs are still used in 3D application only that, but when the multi-hop routing
scenarios in real life, despite the fact that protocol was used, the number of nodes near
there is one application scenario that can the convergence node decreased, which also
be thought of as a 2D environment with decreased the network’s load balance. It is
minimal fluctuation. It is possible to use the inevitable to designate the nodes within
2D routing protocol in a 3D environment. one hop of the SN as the relay nodes in
Under the same circumstances, the LEACH the majority of multi-hop routing protocols.
protocol’s performance in 2D and 3D space is The network load imbalance gets worse
simulated. It is discovered that the lifetime when the number of relay nodes decreases
of the LEACH protocol in 3D space is because the load on the remaining relay
decreased by roughly 21% when compared nodes increases. It is critical to develop
to the LEACH protocol in 2D space. It is an energy optimization protocol for load
established that the traditional protocol’s balancing in order to address the issues
2D approximation of the environment of energy consumption and load imbalance
is illogical. Nodes in a 3D environment in 3DWSNs.
typically use more energy than those in
a 2D environment because of issues with The time for each node to broadcast
the 3D environment’s structure. It is easy CH claim information is computed based
to demonstrate that SN is at the center of on its residual energy and the average
the network if there are an equal number distance between it and surrounding nodes.
of nodes uniformly distributed in the sphere It is preferred for nodes to be CH if they
with radius R and in the circular plane area. have a small average distance from the
In both environments, it is not difficult to surrounding nodes and a relatively high
find through the formula of the circular area residual energy. Between CH and MN, two
S=2πr2 and the formula of sphere volume additional identity nodes are added in an
V=4/3πr3. The proportion of the number of effort to balance the network’s load and
nodes with a distance less than r to the total lower energy consumption. The secondary
number of nodes is reduced from (r/R)2 in 2D SCH serves as the standby CH, and the
space to (r/R)3 in 3D space, that is to say, other is the load transfer node (LTN), which
distributes the communication load among
202 COMPUTER NETWORKS

CH nodes. The CH and SCH nodes are comparatively straightforward because it

generated in the order specified during the was the first to apply clustering to a WSN.
initial clustering process. The SCH nodes Even so, the introduction of the clustering
immediately take on the role of the new concept significantly lowers the single-hop
CH during the ensuing clustering process, distance between any two nodes in the
and only new SCH nodes are created. This entire network. By creating a random number
protocol generates routing tables by using and establishing the cluster probability, the
SN, which has powerful computing power, LEACH protocol chooses CH. To prevent a
in place of OD. Through directed flooding, node from acting as the CH continuously,
the CH node notifies SN of its location and the nodes that have served as CHs are
remaining energy during the network’s unable to run for the CH positions in the
initial clustering. Each CH’s routing table following round. The CHs are distributed
is created by the SN. In accordance with randomly by the LEACH protocol, and it
the routing table, the CH node adds data is not possible to influence or direct this
information to the SN node. Consequently, process. Only fuzzy control, which modifies
when clustering for the first time in the the clustering probability, can limit the total
route construction method, CH nodes number of CHs. This can easily result in a
only need to let SN nodes know where situation where the MN of a particular CH
they are. This greatly lowers the energy is excessively high, disrupting the network’s
consumption during the route construction load balance.
process because the newly generated CH in
the subsequent process only needs to wait Since LEACH cannot reasonably choose
for the routing table created by SN nodes. CHs at random, a new clustering algorithm
SN uses the fragile collection ant colony is developed after LEACH. The hybrid
algorithm to create the routing table that energy-efficient distributed (HEED) [4]
corresponds to each CH. We propose a load protocol represents the clustering technique
balancing routing algorithm for WSNs called used to calculate CHs. The HEED protocol
the fragile collection ant colony algorithm. takes into account the impact of both the
We include the fragile coefficient of every nodes’ residual energy and the WSN’s
edge, based on the ant colony routing overall network energy. The selection of
algorithm, so that the ant takes into account CH nodes goes from completely equal
the energy of each node along the path in probability to unequal probability random,
addition to the path’s overall length. We as the nodes with larger residual energy
compare various routing protocols under have a higher probability of becoming CH.
the network energy consumption model The remaining energy of nodes influences
and the 3D sphere wireless sensor model. the choice of CHs to some degree. However,
The protocol performs better than other when calculating the clustering probability,
protocols in terms of load balancing and the average energy of all the nodes in the
network lifetime, according to the results. network is taken into account. All other
nodes must provide energy information
to nodes. Consequently, nodes’ energy
5.2.3.1. Work Process consumption rises. Energy consumption is
higher in large-scale networks or networks
The LEACH protocol is unavoidable when it
with a high node density. Nonetheless, it
comes to WSNs. The clustering mechanism
makes sense to calculate the likelihood of
is proposed by the LEACH protocol. The CH
nodes turning into CHs using their residual
selection process in the LEACH protocol is
energy.
 NETWORKING FOR CLUSTERS 203
There are additional clustering clustering, the CH gathers and transmits
techniques. For instance, the dynamic node data to the SN; if the cluster fails right
hierarchical clustering in 3D WSNs (3DHCP) away, the network will initiate a fresh round
energy-optimization clustering routing of CH selection. Frequent replacement of
protocol uses the clustering technique the CH prevents the “hot point” issue that
of first determining the ideal number of arises when a node serves as CH for an
clusters in the network. The network is extended period of time; however, excessive
divided into the ideal number of clusters, clustering also increases total energy
and then appropriate nodes are identified consumption and shortens network life.
in each region to become CHs. In this way, Next, we conduct experiments to ascertain
the network’s overall energy consumption this protocol’s ideal clustering period.
is decreased, and its load balance is
improved. However, the time required for In actuality, building multi-hop routing
this clustering method is lengthy, and the for some of the existing multi-hop WSN
size of each cluster cannot be guaranteed. protocols is not too difficult. For instance,
Clusters in the network with a wide size in order to guarantee data transmission in
gap are simple to form. the SN direction, the sending CH chooses
the relay node when the EEUC protocol
There are two types of routing builds the routing path by measuring the
protocols: single-hop and multi-hop. The distance between the candidate CH and
aforementioned LEACH, HEED, and 3DHCP itself. Through flooding, the Advanced Zonal
protocols are examples of single-hop routing Stable Election Protocol (AZ-SEP) protocol
protocols, and situations where there are connects CHs located far from the SN to the
few nodes and limited network coverage SN. However, in recent years, researchers
are typical application scenarios for them. have discovered that the network’s load
The single-hop routing protocol will use a balance and energy consumption will
lot more energy if the network coverage be impacted by the routing system’s
is widespread. Using a multi-hop routing construction. As a result, numerous novel
protocol makes more sense when there approaches to route building are put forth.
is extensive network coverage. Through For instance, the Unequal Clustering Routing
the energy-efficient uneven clustering Protocol (UCNPD) takes into account energy
(EEUC) protocol, the nodes have multi- balancing based on network partition and
hop communication with the SN. Data distance and expands nodes’ sensing
is directly transferred to the SN by CHs range while increasing the routing path’s
with communication capabilities after the consideration factors when establishing
clustering phase. Other CHs utilize a greedy communication between clusters. In the
algorithm to determine their next-hop by building inter-cluster routing stage, UCNPD
calculating the distance from other nodes, adopts a reasonable approach to increase
on the assumption that data forwarding the sensing range of nodes and uses the
to the SN completes the communication. nodes’ remaining energy as the deciding
In order to balance the network’s load factor for next-hop selection. Nonetheless,
and lower energy consumption, the EEUC the growth of.
protocol uses multi-hop and non-uniform
clustering. Additionally, the route construction
process gradually applies the sophisticated
However, it employs the “use and path planning algorithm. Using an enhanced
throws” clustering technique. Following ant colony algorithm (PEG-ant) protocol,
204 COMPUTER NETWORKS

PEGASIS in WSN creates a routing chain connecting every node by

applying the ant colony algorithm. The PEGASIS protocol in the 3D
application environment is enhanced by the PEG-GA (PEGASIS in
3DWSN) protocol, which links nodes into the chain using a genetic
algorithm rather than a greedy one to produce the node chain with
the shortest length.
The ant colony algorithm is used in the adaptive periodic
threshold-sensitive energy-efficient sensor network (APTEEN)
protocol to create multi-hop routing. The network’s performance
has somewhat improved as a result of these protocols. However,
in order to use the routing algorithm to build routing, the OD must
have a large amount of computing and storage capacity. Additionally,
building inter-cluster routing will require more energy to run. Since
the node is widely dispersed throughout the network, an increase
Keyword in OD costs will result in a significant rise in the network’s overall
cost. As a result, we suggest a way to create routing paths using
Path planning
algorithms are SN without raising the price of common nodes.
used in various
fields such as The ant colony algorithm, a well-developed path search algorithm,
robotics, computer can be used to create routing paths since it is frequently used to
graphics, video resolve point-to-point path planning and the traveling salesman
games, and
navigation systems problem (TSP). To better adapt to the environment of WSNs, there
to find an optimal are a few differences between the application conditions of the ant
path from a starting colony algorithm and WSNs that need to be addressed.
point to a goal while
avoiding obstacles
or considering Nodes in a cluster determine how long it will take to send cluster
constraints. head claim information by calculating their distance and relative
energy from other nodes in the cluster. To balance the network
load and lower the energy consumption of the network routing
path construction process, we have introduced two new roles, SCH
and LTN, to the cluster. Regarding the creation of routing paths,
we suggest a novel approach that shifts the responsibility for path
construction from OD to SN. According to simulation results, the
suggested cluster routing protocol improves network load balance
and efficiently lowers network energy consumption.
This makes the following main contributions.
• The reasons for the performance degradation of the two-
dimensional wireless sensor network routing protocol in
three-dimensional wireless sensor networks are analyzed.
• A new model for measuring the fitness value of the cluster
head is proposed. Considering the relative distance between
the nodes and the surrounding nodes, the nodes with a
smaller average distance from the surrounding nodes have
a higher probability to become cluster heads.
 NETWORKING FOR CLUSTERS 205
• The SN replaces the OD and cooperates with the cluster head update mechanism
to generate a routing table corresponding to each cluster head.
• A new path-finding algorithm based on the ant colony algorithm is proposed,
which is more suitable for wireless sensor network routing construction. The
algorithm not only considers the total length of the generated path but also
the energy of the nodes that the path passes through.

5.2.4. Prepare Cluster Networking Infrastructure

This part describes the cluster network preparation required for the installation of an
Exasol cluster.

5.2.4.1. Setup a Basic Network

A public network (CSCN) and a cluster-internal network (CICN) are the minimum
components of the basic network configuration. Every node also features an interface
for out-of-band management, or lights out management.

Figure 5.12. Basic network setup (Source: Exasol, https://docs.exasol.com/db/7.1/resource/images/

administration/on%20premise/installation/network_simple_1_844x410.png).

5.2.4.2. Private Network (CICN – Cluster Internal Communication Network)

Create a dedicated and isolated VLAN (for example, called “VLAN CICN”) and connect
a NIC of each cluster member to it (license node and all data nodes). For the CICN,
configure the following:
• Disable STP (Spanning Tree Protocol)
• Disable EEE (Energy Efficient Ethernet)
• (Optional) Disable Flow-Control to optimize performance for less network latency.
206 COMPUTER NETWORKS

The CICN VLAN must not be routed.

• CICN IP Address Numbering: The CICN is an IPv4 Class
A network by default. Using cluster-internal DHCP, IP
addresses are assigned. The Number setting in EXA
operation configuration refers to the final octet of the node’s
IP address. For instance, the node’s private IP address is
27.1 if its internal number is 15. 0:15. The Node ID and
Internal ID are other names for this number.
5.2.4.3. Public Network (CSCN – Client Side Communication Network)
You can set up a public network in two ways:
• Create a dedicated VLAN (for example, called “VLAN CSCN”)
and connect an NIC of each cluster member to it (license
node and all data nodes).
• Use an existing VLAN and connect an NIC of each cluster
member to it (license node and all data nodes).
5.2.4.4. CSCN IP Address Numbering
The CSCN may be a class C private network or an IPv4 class A
network. 10.50 is the IP address range by default. 0 points out of
24. The External Number setting is used to find the last octet of the
Keyword node’s IP address when configuring nodes in EXA operation. The
last octet of the public network IP address is modified to include
Cluster internal
communication the External Number that was assigned to the node.
network refers to
the infrastructure Example. Client-Side Network Address + External Number =
that enables Client-Side Node IP Address
communication
among the nodes If you plan to use EXA2EXA for data transfer, the public IP
within a cluster
computing addresses of the nodes must be a continuous, consecutive address
environment. In a block.
cluster, multiple
interconnected
computers (nodes) Public Network Network Mask External Number Node IP
work together Access Address
to perform a
task or execute a 10.70.0.50 255.255.255.0 34 10.70.0.84
computation. (24)
10.70.0.70 255.255.255.0 1 10.70.0.71
(24)

Source: Exasol

5.2.4.5. Setup Additional Networks (Optional)

Depending on the need for link bonding or network fail-safety,
there are two approaches to implement more network interfaces.
 NETWORKING FOR CLUSTERS 207
Communication between clients and cluster members can be facilitated by implementing
either of these two approaches. For the cluster internal communication network (CICN),
both approaches are shown in the example that follows. Access to every CICN network
is required for the license node. If the license node has fewer physical network interfaces
than the data nodes, add VLAN tagging (802.1q) to the CICN interface of the license
server.

5.2.4.6. Private Failover Interface (For Network Fail Safety)

Figure 5.13. Add 3rd NIC of each cluster member to existing VLAN ‘CICN’ (Source: Exasol,
https://docs.exasol.com/db/7.1/resource/images/administration/on%20premise/installation/
network_failover_1_799x392.png).

5.2.4.7. Private Bonding Interface (for link bonding)

Create a new dedicated and isolated VLAN ‘CICN_2’:
• Add 3rd NIC of each cluster member to it
• Disable Flow-Control
• Disable STP (Spanning Tree Protocol)
• Disable EEE (Energy Efficient Ethernet)
For multiple internal networks, each needs to have different IP address numbering.
For example:
• Private1 27.1.0.0/16
• Private2 27.65.0.0/16
• Private3 27.66.0.0/16
Internal traffic must not be rooted.
208 COMPUTER NETWORKS

Figure 5.14. Create a new dedicated and isolated VLAN (Source: Exasol, https://docs.exasol.com/
db/7.1/resource/images/administration/on%20premise/installation/network_bonding_1_810x378.
png).
5.2.4.8. Network Switch Configuration • Phase 2: AES 256 / SHA1 / PFS-
Group 5: MODP 1536; Lifetime: 3600
The following features must be disabled on seconds
the switch ports connected to the cluster
For maintenance tasks, the following
to maximize network performance:
service/ports need to be accessible, at
• IEE802.3az: Energy-Efficient least on the cluster’s public and LOM IP
Ethernet (EEE) (“no interface A1- addresses:
A24 energy-efficient-ethernet”)
• TCP/20 (SSH)
• IEEE 802.3x: Flow-Control (“no
• TCP/22 (Cluster-SSH)
interface A1-A24 flow-control”).
• TCP/443 (HTTPS access to
5.2.4.9. Configure VPN Tunnel for Exasol EXAoperation and the LOM web
Support interface)
• ICMP ECHO REQUEST/REPLY
If you opt for Exasol to provide support (ping)
via VPN, you will need to configure a site-
to-site VPN tunnel. The proposed setup is • UDP/123 (NTP)
an IPSec tunnel with the following phases: If you mandate Exasol to monitor
clusters or to perform incident management,
• Phase 1: AES 256 / SHA1 / DH- additional ports will be required.
Group 5: MODP 1536; Lifetime: 7800
seconds

Did you get it?

1. Design network solutions by selecting appropriate clustering protocols based on specific
requirements and constraints.
2. Examine the functionalities and objectives of clustering and routing protocols in WSNs.
 NETWORKING FOR CLUSTERS 209

5.3. NETWORKS AND

INTER-CONNECTION/
SWITCHING DEVICES
→Learning Objectives

• Define the role and function of a Network Interface Card (NIC) in computer
networking
• Define the purpose and functionality of a repeater in networking
• Explain the role of a hub in network connectivity and data distribution
• Analyze the role of bridges in improving network performance, reducing collisions,
and enhancing security
• Describe how switches operate at the data link layer to forward data packets
based on MAC addresses
• Design and implement router configurations to optimize network routing, security,
and performance

A modern network needs a lot of interconnection devices, ranging from the interface
that connects a single computer to other computers through a phone line or LAN cable
to the big, intricate switching devices that link two or more complete networks. The
following is a list of the primary types of interconnection devices used in computer
networks.

5.3.1. Network Interface Card (NIC)

A network interface card (NIC) must be installed on any device connected to a network
that needs to send and receive data. They are often installed into one of the expansion
slots on the computer in the same manner as a sound or graphics card, and are also
referred to as network adapters. A transceiver—a transmitter and receiver combined—is
part of the NIC. A network device can send and receive data over the transmission medium
with the help of a transceiver. Every NIC is manufactured with a distinct 48-bit Media
Access Control (MAC) address burned into its ROM. An Organizationally Unique Identifier
210 COMPUTER NETWORKS

(OUI), a block code that uniquely identifies Operationally, a repeater is at the physical
the manufacturer of NICs, is made up of layer. Its task is to prolong the time that
the first 24 bits. The Institute of Electrical a signal can be transmitted over the same
and Electronics Engineers (IEEE) oversees network by regenerating the signal over it
the issuance of OUIs to organizations. before it gets too weak or distorted. It is
The manufacturer’s sequential number is imperative to acknowledge that repeaters
comprised of the final 24 bits. The MAC do not enhance the signal strength. They
address, which uniquely identifies the replicate the signal bit by bit and regenerate
network adapter, is also referred to as a it at the original strength when it starts to
hardware address or physical address. fade. This gadget has two ports.
It is used by many data link layer
communications protocols, including 5.3.3. Hub
Ethernet, the 802.11 wireless protocol,
An Ethernet network uses hubs. Every
and Bluetooth. The use of a 48-bit address
port on the hub retransmits a signal that
allows for 248 (281,474,976,710,656) unique
is received at any other port. Hub-using
addresses. A MAC address is usually shown
network segments are frequently referred
in hexadecimal format, with each octet
to as having a star topology, where the hub
separated by a dash or colon, for example:
serves as the star’s wiring center.
00-90-47-03-B6-C4.
Because each network device has a
Did you know? separate connection to the hub, using a
The NIC is both a physical layer and data link layer
hub offers a certain level of fault tolerance
device, as it provides physical access to a networking as only one device is impacted by a failed
medium and, for IEEE 802 and similar networks, connection. Additionally, it is simpler to
provides a low-level addressing system through the use expand the network because a single hub—
of MAC addresses that are uniquely assigned to network which is frequently linked to a network
interfaces.
backbone—can add a large number of new
devices. Hubs may be passive or active. An
active hub regenerates incoming frames
5.3.2. Repeater prior to retransmitting them and has its own
power supply. Each output port can link a
Signal strength will decrease (attenuation)
channel up to 100 meters (the maximum
as it passes through a transmission medium.
permitted length for twisted pair cables)
A repeater is a type of network device that
because signals are regenerated. Passive
lacks intelligence. It receives a signal on one
hubs don’t generate the signal; they just
port, regenerates it, and then retransmits
relay it.
it on all of its remaining ports. Repeaters
can join two network segments together
to increase a network’s length (but not 5.3.4. Bridge
its capacity). Repeaters are not useful for
Bridges are useful for connecting multiple
connecting network segments that use
LAN segments of the same kind (e.g.,
distinct network access methods or for
Token-Ring to Token-Ring or Ethernet to
expanding a network beyond the bounds
Ethernet). Similar to repeaters, bridges can
of its underlying architecture. However, they
increase a network’s length, but because
may be able to link segments of bridges with
each port on a bridge has a unique MAC
varying data rates and various media types.
 NETWORKING FOR CLUSTERS 211
address, they can also increase a network’s capacity. In an Ethernet
network, bridges that are powered on begin by analyzing the
source addresses of incoming frames from all connected network
segments in order to determine the topology of the network. They
accumulate a routing table over time. The bridge does not need
to move an incoming frame to another network segment unless
the source and the destination are on separate network segments.
The bridge must be able to identify which segment the destination
device is on if the source and destination are on separate segments.
The bridge monitors all traffic on the segments it connects,
cross-referencing each frame’s source and destination addresses
against its routing table. The routing table is empty when the bridge
first starts up, but as data is sent back and forth, the bridge adds Remember
each incoming frame’s source MAC address to the table and links
it to the port on which the frame arrives. The bridge quickly builds Managed
hubs give
a comprehensive image of the network topology in this way. An administrators
incoming frame will be forwarded to all attached segments if the the ability to
bridge is unsure of its destination segment, with the exception of remotely enable
the segment from which it originated. By segregating intra-segment or disable specific
ports, and
traffic and acting as a filter, bridges reduce the volume of traffic on
intelligent hubs
individual segments. This can significantly improve reaction times. have the ability
to automatically
Ethernet bridges are considered to operate at the data link layer close ports if a
of the OSI Reference Model because they decide whether or not to predetermined
forward frames based on the destination MAC address. Although threshold of
they effectively isolate intrasegment traffic, lowering overall network packet errors is
exceeded.
traffic and speeding up network response times, Ethernet bridges
are sometimes referred to as transparent due to the transparency
of both their existence and operation to network users.

5.3.4.1. Bridging Loops

When more bridges are added to the network to offer a degree
of redundancy and fault tolerance by providing alternate paths
through the network, a potential issue could occur. Loops may
arise if there are multiple paths connecting any two segments at
the same time. Bridges can be easily confused in the very basic
example shown in the diagram below. For instance, these bridges
would continuously forward a broadcast message from A, consuming
valuable network bandwidth and preventing the transmission of
other frames on both segments.
By putting redundant links into standby mode, which can
be activated in the case of a primary link failure and offer a
new path through the network, a Spanning Tree Algorithm (STA)
can remove loops. The algorithm needs to be dynamic in order
212 COMPUTER NETWORKS

for the bridges to be able to recognize the bridge itself. Network administrators
changes in topology and generate a have the ability to assign an arbitrary cost
new spanning tree accordingly. When to each bridge port on the network. The
the bridges are initially powered on and algorithm can then use this cost to decide
whenever a change in the network topology which ports are put into standby mode and
is noticed, the spanning-tree calculation which are left open, based on which ports
takes place. Configuration messages are have the lowest cost. A default value is
used to facilitate communication between applied if no cost value is specified. The
the bridges and are sent back and forth network with five segments that redundant
at regular intervals (usually every few bridges have caused loops within is depicted
seconds). The other bridges will rerun the in the diagram below.
algorithm if a bridge fails because they
will notice that its configuration messages The root bridge is chosen initially in a
are missing. spanning-tree computation; this is typically
the bridge with the lowest identifier. For
the purpose of clarity, the simplified bridge
5.3.4.2. The Spanning Tree Algorithm identifiers (B1-B5) are used in the diagrams
(Ethernet) on this page, and B1 will be the root bridge.
Next, one needs to identify a root port (R)
The spanning tree technique disables loops
on every other bridge. This is the port that
in a network by putting bridge ports into a
has the lowest root path cost to reach the
standby (blocking) state that, in an active
root bridge (the cost of each bridge port
state, would result in loops. In the event
that must be crossed on a specific root
that a link fails to offer an alternative path
path will determine this). Lastly, for every
across the network, a standby port can be
LAN segment, designated bridges and their
reactivated. The spanning tree algorithm
designated ports are identified. The bridge
builds a loop-free topology using graph
on the segment that offers the root path at
theory as a foundation. According to graph
the lowest cost is designated. The port that
theory, a spanning tree of edges keeps a
links a segment to its designated bridge is
connected graph connected but free of
known as the designated port. The bridge
loops, given any connected graph made up
with the lowest bridge identifier number
of nodes and edges joining pairs of nodes.
is used when two or more bridges have
Every bridge represents an edge, and the same root path cost. All but one of
every LAN segment represents a node. the bridges attached to each LAN segment
This makes it possible for the network’s are removed using the STA, eliminating
bridges to generate a spanning tree using loops while maintaining connectivity. The
a straightforward algorithm. Bridges can outcome of applying the STA to the network
register a change in network topology and that was previously depicted is displayed in
obtain a new spanning tree by rerunning the the following updated diagram. The ports
STA, demonstrating the dynamic nature of company is in the STA’s list.
the algorithm. According to the algorithm,
When a change in the network topology
every bridge must have a unique identity, as
is detected or the bridge is powered on,
does every port on every bridge. Each port is
the spanning-tree calculation takes place.
typically identified by its MAC address, and
Bridges must communicate with one
the algorithm uses one of the bridge ports—
another in order for this to occur, and bridge
prefixed by a two-byte priority—to identify
protocol data units (BPDUs) are used to
 NETWORKING FOR CLUSTERS 213
send configuration messages. The bridge and port identifiers of the
sending bridge are contained in a configuration message, which
also identifies the bridge that the sender believes to be the root
bridge.
It also includes the age of the data in the configuration message
and the root path cost from the sending bridge to the root bridge.
Every few seconds, on average, bridges exchange configuration
messages. In the event of a bridge failure, which results in a
topology change, neighboring bridges will detect this and send
topology change messages to start a spanning-tree recalculation.
Keyword
Bridge Protocol
5.3.4.3. Frame Format Data Units
(BPDUs) are
The format of the IEEE 802.1d configuration message is illustrated frames that contain
information about
in Figure 5.15. The fields of the bridge configuration message are the spanning tree
described below. protocol (STP).

• Protocol identifier – contains the value zero.

• Version – contains the value zero.
• Message type – contains the value zero
• Flags – only the first two bits are used. The topology-change
bit , if set, signals a topology change, and the topology-
change acknowledgment bit , if set, acknowledges receipt of
a configuration message with the topology-change bit set.
• Root ID – identifies the root bridge using its 2-byte priority
followed by its 6-byte ID.
• Root path cost – the root path cost from the bridge sending
the configuration message to the root bridge.
• Bridge ID – identifies the bridge sending the message using
its 2-byte priority followed by its 6-byte ID.
• Port ID – identifies the port from which the configuration
message was sent.
• Message age – indicates when the configuration message
should be deleted.
• Maximum age – contains the value zero.
• Hello time – indicates the time period between root bridge
configuration messages.
• Forward delay – the time bridges should wait before
transitioning to a new state after a topology change.
Topology-change messages are only 4 bytes long and consist
of the protocol identifier field , the version field , and the message
type field which contains the value 128.
214 COMPUTER NETWORKS

Figure 5.15. Bridge configuration message format (Source: Research gate, https://www.researchgate.
net/profile/Soufiene-Djahel-2/publication/280317385/figure/fig3/AS:287057567076355@1445451265
758/Format-of-the-new-RREQ-packet.png).

5.3.4.4. Route Bridging (Token Ring)

The source-route bridging (SRB) algorithm, which IBM created for Token Ring networks,
gets its name from the fact that the sending device inserts routing information into each
and every inter-segment frame. forwards frames by using the routing data contained in
the frame as a guide. Below is an example of a basic source-route bridging network.
Let us assume that computer A wishes to transmit a frame to computer B. Initially,
A does not know if B is located on the same ring or a separate one. Therefore, A
sends a test frame. If B does not acknowledge the test frame, A assumes that B is
on a different segment. Subsequently, A sends out an explorer frame. The explorer
frame is forwarded to any additional segments connected to each bridge that receives
it (in this case, B1 and B4). As the explorer frame traverses the network, each bridge
appends route information to it. Upon reaching B, each explorer frame receives a unique
reply that is routed based on the accumulated route information. After receiving the
replies, A must choose one of the listed routes for future transmissions to B, following
predefined criteria.

Figure 5.16. A simple source-route bridging network (Source: https://www.vikramuniv.ac.in/files/

academic/e-Resources2020-21/engg2020-21/BE_8_SEM_EL_CN-DLL_SWITCHING-AMIT_THAKUR.
pdf).
 NETWORKING FOR CLUSTERS 215
Typically, the route provided in the first for route discovery.
reply is selected, although other factors such • Length (5 bits) – specifies the total
as the maximum frame size allowed and the length of the routing information
number of hops on a specific route may field in bytes.
also influence the decision. Once a route
• Direction bit – indicates the
is chosen, it is inserted into the routing
direction in which the route is
information field (RIF) of frames destined
being traversed, and determines
for B. Only frames addressed to devices
whether bridges read the ring and
on different segments contain a routing
bridge number combinations in the
information field. Indicating the presence
route designators from right to left
of routing information in the frame is done
(forward) or left to right (reverse).
by setting the most significant (leftmost)
bit in the Source Address field. This bit • Largest frame (6 bits) – indicates
is referred to as the routing information the largest frame size that can be
indicator (RII) bit. handled by the specified route.
• Reserved bit – not used
5.3.4.5. Frame Format • Routing descriptor fields – contains
the following subfields:
The location of the routing information field
– Ring number (12 bits) –
within a Token Ring data/command frame
identifies a ring segment
is shown below, together with its detailed
(must be unique within the
structure.
network)
The routing information field consists of – Bridge number (4 bits) –
a single routing control field and a number identifies a bridge (only has
of routing descriptor fields. These fields and to be unique if the bridge
their subfields are described below. it identifies is parallel with
another bridge connecting
• Routing control field – contains the
two rings).
following subfields:
A maximum of 14 routing descriptor
– Routing type (3 bits) – indicates
fields may be present in the routing
one of three possible types of
information field. Every route consists of
routing control:
a series of bridge and ring numbers that
– Specifically routed – used when alternate and begin and end with a ring
the source node supplies the number. For every explorer frame that a
route in the RIF’s header. bridge receives, it adds its bridge number
– All paths explorer – used to and the number of the ring that the frame is
find a remote node. Route forwarded onto (the first bridge additionally
information is collected as the adds the ring number of the ring that it
frame traverses the network. receives the frame from, and the last bridge
– Spanning-tree explorer – used number always equals zero).
to find a remote node. Only
bridges in the spanning tree
5.3.5. Switch
forward the frame, reducing
the number of frames used Hubs and bridges in Local Area Networks
have been superseded by the relatively
216 COMPUTER NETWORKS

new network device known as switches. An also known as an intermediate system at

internal address table is used by a switch times, must be capable of determining the
to route incoming data frames to the port optimal route to forward incoming packets,
that corresponds to their final MAC address. or datagrams, to the destination network.
Switches are useful for interconnecting Routers gather information about the
different network segments or several end- networks they are linked to and can share
user devices, like workstations. A workgroup this information with other routers on the
switch is a term frequently used to describe same network. The accumulated data, which
a switch that links end-user devices. Each includes the routing information itself and
connected device is essentially given its the current status of various network links,
own network segment by the dedicated is stored in the router’s internal routing
full-duplex links that switches provide. This table. Through specialized routing protocols,
greatly lowers the amount of intra- and routers exchange this routing data.
inter-segment collisions.
In many cases, hosts or end-systems
A floor switch is a higher-level switch refer to computers and other end-user
that connects several network segments devices connected to networks that form
that are on the same floor of a building an internetwork. A network host will direct
(or on the same campus) and each has a datagram to its local router (also known
its own workgroup switch. Floor switches, as the default gateway) if it cannot route
workgroup switches, routers, and other the datagram to a host on another network.
high-level network devices are frequently As a datagram travels from an end-system
connected by far more powerful switches. on the source network to an end-system on
These devices, which go by the name “core the destination network, it may traverse
switches” frequently, ought to be able to multiple networks and, consequently,
handle the amount of traffic that moves multiple routers. Each intermediate router
through the network. makes a determination regarding the best
next hop. Routers serve as the central
element of packet-switching networks. The
5.3.6. Router process by which the router forwards the
A sophisticated device is required to incoming datagram to one of its output ports
manage the flow of traffic between multiple in this manner is referred to as switching.
interconnected networks with different Routers utilize the IP address contained
network protocols and architectures in a in a datagram to ascertain the destination
network environment. This type of device, network address, unlike bridges and
commonly referred to as a router but switches, which focus on MAC addresses.

Did you get it?

1. Explain the role of a hub in network connectivity and data distribution
2. Design and implement router configurations to optimize network routing, security, and
performance
 NETWORKING FOR CLUSTERS 217

A CLOSER LOOK
Cluster networks are groups of high-performance computing (HPC), GPU, or optimized
instances that are connected with a high-bandwidth, ultra-low-latency network. Each
node in the cluster is a bare metal machine located in close physical proximity to the
other nodes. A remote direct memory access (RDMA) network between nodes provides
latency as low as single-digit microseconds, comparable to on-premises HPC clusters.
Cluster networks are designed for highly demanding parallel computing workloads.
For example:
• Computational fluid dynamics simulations for automotive or aerospace modeling
• Financial modeling and risk analysis
• Biomedical simulations
• Trajectory analysis and design for space exploration
• Artificial intelligence and big data workloads
Oracle Cloud Infrastructure offers two types of cluster networks. In both cases,
the networks are groups of bare metal instances that are connected with an ultra-low
latency network.
• Compute clusters allow you to manage instances in the cluster individually.
When you create a compute cluster, you create an empty RDMA network
group. After the group is created, you can add instances to the group or delete
instances from the group. If you want to manage instances in the RDMA network
independently of each other or use different types of instances in the network
group, use compute clusters.
• Cluster networks with instance pools enable you to use instance pools to
manage groups of identical instances in the RDMA network group. If you want
predictable capacity for a specific number of identical instances that are managed
as a group, use cluster networks with instance pools.
218 COMPUTER NETWORKS

SUMMARY
• A cluster network is an ultra-low latency, high-bandwidth network that connects a
pool of GPU-optimized instances or high-performance computing (HPC) instances.
• The networks whose connectivity fluctuates frequently (e.g., mobile networks) or
are large with many configuration parameters work better with self-organizing
controlled structures. Because nodes construct and maintain them, they offer
greater network availability, a quicker reaction to state changes, and a lower
chance of configuration errors.
• Broadcast environments with multiple access utilize this architecture. Node
clusters are organized to control transmissions without contention, reducing
interference..
• Large tactical networks are the intended application and deployment context
for Network Near-Term Digital Radio (NTDR) networking. This is one of the
backbone formation clustering techniques.
• The cluster heads in the NTDR network are responsible for keeping the backbone
stable, so they monitor it and share information about any changes with one
another.
• The quasi-hierarchical category includes the majority of hierarchical cluster-
based routing systems. Moreover, the distance vector routing technique forms
the basis for nearly all quasi-hierarchical routing schemes.
• A node strives to create routes that align with the users’ service requests and
the network’s existing capabilities using this link-state information.
• Techniques for clustering can be decentralized or centralized. Each sensor node
must transmit its unique data, including its location and energy level, to the
central base station to utilize centralized clustering algorithms.

MULTIPLE CHOICE QUESTIONS

1. What is a primary benefit of cluster networking?
a. Reduced hardware costs
b. Increased latency
c. Limited scalability
d. Single point of failure
2. Which of the following accurately describes a cluster network?
a. A network where all devices communicate through a central server
b. A network where multiple interconnected servers work together to provide
services
c. A network topology with a linear connection between devices
d. A network consisting of isolated, non-communicating segments
 NETWORKING FOR CLUSTERS 219
3. In cluster networking, what is the purpose of load balancing?
a. To increase latency
b. To overload certain servers for better performance
c. To evenly distribute workloads across multiple servers
d. To reduce the number of servers in the cluster
4. Which of the following is a characteristic of high availability in cluster networking?
Single point of failure
a. Redundancy and failover mechanisms
b. Limited scalability
c. Increased latency
d. Single point of failure
5. What role does fault tolerance play in cluster networking?
a. It increases the chances of system failures
b. It ensures that the network operates at maximum capacity
c. It allows the network to continue operating even if some components fail
d. It reduces the need for load balancing

REVIEW QUESTIONS
1. What is the role of routing in cluster-based networks?
2. How can clustering protocols be classified?
3. What are the considerations in clustering and routing protocol design for wireless
sensor networks?
4. What functions does a Network Interface Card (NIC) serve?
5. How does the Spanning Tree Algorithm apply to Ethernet networks?

Answer to Multiple Choice Questions

1. (a) 2. (b) 3. (c) 4. (b) 5. (c)

REFERENCES
1. Charles, E., (2001). Perkins: Chapter 4. Ad Hoc Networking. Addison Wesley. ISBN
0-201-30976-9.
2. Kalashnikov, D. V., & Mehrotra, S., (2004). Learning Importance of Relationships for
Reference Disambiguation. RESCUE.
3. McDonald, D. W., (2003). Recommending collaboration with social networks: A
comparative evaluation. In: Proceedings of the SIGCHI Conference on Human Factors
in Computing Systems (pp. 593–600). ACM Press, Ft. Lauderdale, Florida, USA.
220 COMPUTER NETWORKS

4. Tho, Q. T., Hui, S. C., & Fong, A. C. M., (2003). A web mining approach for finding
expertise in research areas. In: Cyberworlds, 2003; Proceedings. 2003 International
Conference (pp.310–317). Oxford University Press.
5. Bathelt, H., & Glückler, J. (2011). The Relational Economy: Geographies of Knowing
and Learning. Oxford University Press.
6. Ketels, C. (2013). Cluster Reinvention and Regional Economic Development. In M.
Porter, C. Ketels, & M. Delgado (Eds.), Competitiveness in the Global Economy: A
Multi-Disciplinary Perspective (pp. 235-258). Harvard Business School Publishing.
7. Martin, R., & Sunley, P. (2003). Deconstructing clusters: Chaotic concept or policy
panacea? Journal of Economic Geography, 3(1), 5-35.
CHAPTER
Distributed
6 Network Systems

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Elaborate the client-server model in a distributed

computing system
• Examine the extensions to the client-server model
• Discuss on service discovery
• Explain the characteristics of a distributed system
222 COMPUTER NETWORKS

INTRODUCTORY EXAMPLE

ISTRIBUTED NETWORK IN MODERN COMPUTING

Distributed network systems are a fundamental architectural approach in modern
computing that distribute tasks or resources among multiple nodes, enabling scalability,
fault tolerance, and efficient resource utilization. These systems are characterized by
decentralization, where no single node or entity controls the entire network, and nodes
communicate and collaborate autonomously to achieve common goals. One prominent
example of a distributed network system is blockchain technology, which underpins
cryptocurrencies like Bitcoin and Ethereum. In a blockchain network, transactions are
recorded and validated by a decentralized network of nodes, each maintaining a copy
of the ledger. Through cryptographic techniques and consensus algorithms such as
Proof of Work (PoW) or Proof of Stake (PoS), blockchain networks ensure the integrity
and immutability of data without the need for a central authority. This decentralization
enhances security and eliminates single points of failure, making it resilient to attacks
and censorship.
Another example of distributed network systems is content delivery networks (CDNs),
which optimize the delivery of web content to users by strategically distributing it across
a network of servers located in various geographic regions. CDNs cache static content
such as images, videos, and scripts on edge servers closer to end-users, reducing latency
and improving load times. By distributing content across multiple nodes, CDNs can
handle high traffic volumes efficiently, ensuring a smooth and responsive user experience
even during peak demand periods. Companies like Akamai, Cloudflare, and Amazon
CloudFront provide CDN services, leveraging distributed architectures to enhance the
performance, reliability, and security of websites and web applications.
Peer-to-peer (P2P) networks are another type of distributed network systems where
nodes in the network act as both clients and servers, sharing resources or tasks directly
with each other without the need for central coordination. P2P networks are commonly
used for file sharing, with protocols like BitTorrent allowing users to distribute and
download files across a decentralized network of peers. Each peer contributes bandwidth
and storage capacity, enabling efficient distribution of large files while reducing the
load on individual servers. Similarly, communication applications like Skype use P2P
architectures to facilitate direct communication between users, leveraging the collective
resources of network participants to establish connections and transmit data without
relying on centralized servers. Distributed file systems also demonstrate distributed
network systems, providing storage resources across multiple nodes in a network,
typically for enterprise or cloud computing environments. Examples include Google
File System (GFS), Apache Hadoop Distributed File System (HDFS), and Amazon Simple
Storage Service (S3). These distributed file systems partition and replicate data across
multiple nodes, ensuring high availability, durability, and scalability. By distributing data
storage and processing tasks across a network of nodes, distributed file systems can
handle large volumes of data and support parallel processing, enabling efficient data
analysis, backup, and retrieval operations.
INTRODUCTION
A computer network that is dispersed over several networks is known as a distributed
network. This provides a single data communication network that each network can
manage independently or in concert. In addition to facilitating shared communication
among nodes in the network, a distributed network frequently shares processing.
Distributed computing architecture, which divides enterprise IT infrastructure
resources across multiple networks, processors, and intermediary devices, includes
distributed networks. Network management software powers distributed networks by
combining and allocating network bandwidth, controlling access, managing and tracking
data routing, and performing other essential networking tasks.
Different remote users can receive specialized applications from distributed networks
and processing. This implies that numerous people may access an application that is
hosted and run on a single machine. An example of a distributed network is a client/
server computing architecture, in which numerous connected remote users access the
application from various networks, while the server provides the resource.

DISTRIBUTED NETWORK SYSTEMS 223

6.1. THE CLIENT-SERVER
MODEL IN A DISTRIBUTED
COMPUTING SYSTEM
→Learning Objectives

• Understand features and issues of client-server model in distributed computing.

• Explore cooperation mechanisms between clients and servers in client-server
architecture.
• Differentiate types of cooperation in client-server interactions and analyze
chained servers concept.
• Investigate advantages and challenges of deploying and managing multiple
servers in client-server environment.

A distributed computing system is a collection of data and application programs spread

over separate personal computers linked by a communication network. Users access
services by running necessary programs on multiple computers with data stored in
various locations.
Challenges in distributed computing include formalizing progress and managing
processes that may work together or against each other. The client-server model is
suitable for distributed computing, handling distribution issues, explaining computation
processes, and enabling software development. In this model, the client requests a
service, the server provides it, and responds after completing the task.
Printing, email, file services, authentication, naming, database services, and computing
services are among the services that a distributed computing system can offer from
the user’s perspective. Appropriate servers offer these services. Clients compete for
these servers because there are only a limited number of servers (implied by a limited
number of resources on which these servers were implemented).
Figure 6.1 illustrates the relationship between this conceptual framework and its
concrete application. The client and server, as well as the request and response, are

224 COMPUTER NETWORKS

 DISTRIBUTED NETWORK SYSTEMS 225
the main components of the model that are displayed. Two separate computers are
used in this instance to run the client and server processes. By exchanging requests
and answers, they engage in virtual (logical) communication. These two processes send
physical messages to each other in order to accomplish this virtual communication. This
suggests active participation in the service delivery on the part of computer operating
systems and distributed computing systems’ communication networks.

Figure 6.1. The basic client-server model (Source: Wikipedia, https://upload.wikimedia.org/

wikipedia/commons/thumb/c/c9/Client-server-model.svg/1024px-Client-server-model.svg.png).

A more detailed client-server model has three components:

• Service: A service is a software entity that runs on one or more machines.
It provides an abstraction of a set of well-defined operations in response to
applications’ requests.
• Server: A server is an instance of a particular service running on a single machine.
• Client: A client is a software entity that exploits services provided by servers.
A client can but does not have to interface directly with a human user.

6.1.1. Features and Problems of the Client-Server Model

The client-server model’s simplicity, modularity, extensibility, and flexibility are its key
components. A sign of simplicity is when the data flow and the control flow closely
align. A collection of computer operations can be arranged and integrated into a distinct
service to achieve modularity. Additionally, any collection of data that has operations
on it can be arranged into a distinct service. By introducing new services in the form
of additional servers, the client-server model-based distributed computing system can
be readily expanded. It is simple to change or even remove servers that do not meet
user requirements. All that needs to be maintained are the client-server interfaces.
There are three major problems of the client-server model:
• One server has centralized control over all resources, which is the cause of the
first. This implies that the failure of the computer supporting a server results in
the failure of that control element. Such a solution is unacceptable if a server’s
226 COMPUTER NETWORKS

control function is essential to the retrieves the data from the mail
system’s functionality (e.g., as an server. Every kind of computer
authentication server, a file server, goes through this procedure again
or a name server). As a result, the when requesting information from
dependability of all computers and servers. As such, your phone,
devices as well as communication laptop, and even smartwatch can
lines determines the availability serve as clients.
and reliability of an operation that
depends on numerous servers.
• The potential bottleneck caused
by individual servers is the second
issue. As more computers with
potential customers are added to
the system, the issue worsens.
• The third issue arises when a client-
server based system’s performance
is improved by using multiple
implementations of comparable
functions to maintain consistency.
Figure 6.2. Cooperation functionality between
Additionally, this increases the client and server (Source: TechTerms, https://
overall costs of a distributed medium.com/@joshjnunez09/the-client-server-
computing system. relationship-9ac90fadb3d2).

• The Relationship: We can delve

6.1.2. Cooperation between further into the relationship’s
Clients and Servers process now that you have a clearer
It is highly likely that you are familiar grasp of each component. Like in
with the terms “client” and “server” if a lot of other relationships, every
you have worked in the tech industry for component has a distinct function.
any amount of time. Understanding the For example, one or more clients
relationship between the two is essential can receive a function or service
for comprehending many aspects of common from the server. These customers
applications or webpages. The field of are the ones who are requesting
computer science is built on this model. this service. Only when both roles
We will now review the model, examine are performing their duties is this
its significance, as well as its benefits and relationship feasible. A request-
drawbacks. response messaging pattern is the
exchange of messages between the
• The Client: The client is the other two parties. The server is required
half of this partnership. Additionally, to respond to requests sent by the
the client is a computer unto itself. client. There are several benefits
When this computer retrieves data to this kind of relationship. One
from the server, it does so as a benefit is that all the necessary
client. When you use your laptop data can be stored on the server in
to check your emails, for example, one location. This makes granting
your laptop becomes the client and
 DISTRIBUTED NETWORK SYSTEMS 227
authorization and protecting your data much simpler. To
make matters worse, efficient access to the data is possible
even when the server is not physically close to the client.
Lastly, because the client-server model is independent of
all other models, upgrading the modes is simple. One of
the model’s few drawbacks is that if a client sends in
an excessive number of requests, the server may quickly
become overloaded.

6.1.3. Cooperation Type and Chained Server

Users wouldn’t expect top-notch, cost-effective, and reliable services
from a system with just one server and one client. To handle Keyword
multiple clients simultaneously, a single server is necessary. The most Authorization
basic form of cooperation between clients and servers is sharing, is the function of
which reduces system costs and enhances the efficient use of specifying access
rights/privileges
resources. A printing service exemplifies a service that depends on to resources,
this collaboration. In Figure 6.3, a printer server is depicted serving which is related to
n clients connected via a local area network. There are two main general information
security and
ways clients and servers can collaborate in a distributed computing computer
system. In the first scenario, a client requests a temporary service. security, and to
The second involves a customer scheduling multiple calls to be access control in
particular.
directed to a specific serving procedure. This implies that long-term
connections between the client and server are essential.

Figure 6.3. Printing service (a service example) (Source: What-is-my-

computer.com, https://www.what-is-my-computer.com/images/what-is-a-
print-server.jpg).

6.1.4. Multiple Servers

A distributed computing system has the following functions:
• Improve performance through parallel execution of programs
on a cluster (sometimes called a network) of workstations,
228 COMPUTER NETWORKS

• Decrease response time of databases 6.1.4.2. Cooperation in the Systems

through data replication, Supporting a Distributed Database
• Support synchronous distant
Supporting a distributed database has
meetings,
similarities to the semantics of collaboration
• Support cooperative workgroups, between a client and several servers in
• Increase reliability by service a distributed computing system. Every
multiplication, etc. operation carried out on a group of
Numerous servers must contribute to the databases has to be finished successfully
overall application in order for it to carry out in order to commit the transaction. In order
these tasks. This suggests that using several to complete a transaction, a client process
services is necessary. Moreover, it would sends operation requests to the appropriate
necessitate sending multiple simultaneous databases (servers) and then waits for the
requests to different servers in certain operations’ outcomes. Although the client
situations. As the following paragraphs process is capable of performing other tasks,
demonstrate, different applications will call the transaction cannot be committed until
for different semantics in order for clients all database servers have responded. In
and servers to cooperate. this instance, setting up servers on unused
computers is not necessary because servers
that are currently operating on dedicated
6.1.4.1. Cooperation in the Systems computers already exist. The question of
Supporting Parallel Execution who has access to these database servers,
the client process, or another organization
Certain program components may run as
acting on the client’s behalf, remains,
separate processes on different computers
though.
in a distributed computing system that
supports parallel execution. In order to
accomplish this, a process (parent) that 6.1.4.3. Cooperation in the Systems
orchestrates the parallel processing of Supporting a User Application
distinct processes (children) sets them to run
on particular idle computers while it awaits In a distributed computing system that
the computation of results. In this instance, supports a user application, cooperation
there is a one-to-many communication between a client and several servers
pattern where the parent process serves can take on distinct meanings. Finding
as a client and the child processes serve as a database server that houses pertinent
servers. However, until all children return data, gaining access to it, processing it, and
their responses, the parent process comes to printing it are all necessary for this. Three
an end. This example demonstrates that in servers—a database server, a printer server,
order to enhance the collaboration between and a service discovery server—must be
clients and servers, two questions need active in order to support the application.
to be addressed: first, who is in charge of The application software-running client
identifying idle computers that can host process calls each server in turn, one after
servers, and second, who is in charge of the other, and then waits for a response
installing those servers on distant computers before moving on to the next server. Again,
and organizing responses. there’s no need to install servers on unused
PCs in this scenario. But the same problem
still exists, i.e., who is capable of handling
 DISTRIBUTED NETWORK SYSTEMS 229
these database servers, the client procedure, or another organization
acting on the client’s behalf.
Remember
6.1.4.4. Cooperation in the Systems Supporting Mission Critical Distributed
Applications computing
refers to a
Reliability in distributed computing systems requires a different kind system where
processing and
of cooperation between a client and multiple servers, especially when data storage is
supporting mission-critical applications. In this scenario, a request distributed across
made to a collection of servers ought to result in the same replies multiple devices
from each of them. A fault-tolerant transaction-oriented database or or systems,
a redundant computational server on a spacecraft are two examples rather than
being handled
of such a system. In any event, the client will only accept identical by a single
operation results. These illustrations demonstrate how, in an effort to central device.
increase efficiency and dependability, distributed computing systems In a distributed
have shifted from the traditional one-to-one client-server model to system, each
device or system
the one-to-many and chain models. Additionally, the problems found has its own
when debating the client-server model’s one-to-many communication processing
pattern show that certain active entities that are client-server model capabilities and
extensions can have a significant impact on and support client and may also store
server cooperation. These topics are covered in the sections that and manage its
own data.
follow.

Did you get it?

1. What are the mechanisms of cooperation employed in systems supporting
parallel execution?
2. How does cooperation manifest in systems designed to support mission-
critical applications?
230 COMPUTER NETWORKS

6.2. EXTENSIONS TO THE

CLIENT-SERVER MODEL
→Learning Objectives

• Explore the concept of agents and their role in facilitating indirect client-server
cooperation within distributed systems.
• Understand the principles and components of the three-tier client-server
architecture in distributed computing environments.

The client-server model is a foundational architectural design used in computer networking

that facilitates the communication and exchange of data between multiple devices over
a network. In this model, the interaction between computers is structured into two
distinct roles: the client and the server.
At its core, the client-server model operates on a principle of division of labor,
where specific tasks are assigned to each participant. The client, typically a software
application or device, initiates requests for services or resources from another program
or computer known as the server. The server, on the other hand, is a specialized system
designed to provide the requested services or resources to the client.
The interaction between clients and servers follows a request-response pattern.
When a client requires a service or resource, it sends a request to the server. This
request contains information about the desired service or resource, such as the type of
operation to be performed and any necessary parameters. Upon receiving the request,
the server processes it, executes the requested operation, and formulates a response.
This response contains the requested data or confirmation of the completed operation,
which is then sent back to the client.
One of the key advantages of the client-server model is its scalability and flexibility.
By distributing tasks between clients and servers, the model allows for efficient resource
allocation and utilization. Servers can be dedicated to specific functions or services, such
as file storage, database management, or web hosting, while clients can access these
resources as needed without having to manage the underlying infrastructure themselves.
This separation of concerns simplifies system administration and maintenance, as changes
or updates can be made to individual components without affecting the entire network.
 DISTRIBUTED NETWORK SYSTEMS 231
Additionally, the client-server model that handles resource management and
promotes a centralized approach to cooperative servers, mediates disputes
data management and security. Servers between clients and servers, and resolves
act as central repositories for data and heterogeneity problems. A client can send
resources, which can be accessed and direct requests to multiple servers in
managed through controlled channels. This order to explicitly invoke desired servers.
centralized control enables administrators In this scenario, the user application
to enforce access restrictions, implement programmer needs to focus on managing
security protocols, and monitor network server cooperation and communication in
activity more effectively, reducing the risk addition to the application itself. Writing
of unauthorized access or data breaches. communication and resource management
software is costly, time-consuming, and
However, the client-server model is prone to mistakes. The user interface is
not without its limitations. One potential not transparent to the client process, and
drawback is the reliance on a single point of the interface between the server and the
failure—the server. If the server experiences client varies depending on the application.
downtime or becomes overwhelmed with
requests, it can disrupt service for all
connected clients. To mitigate this risk,
redundant servers and load balancing
techniques can be implemented to distribute
workload and ensure high availability.
A study into the involvement of other
entities in the provision of services, an
interface between a client and server, and
the behavior of a client after sending a
request to a server can specify the need Figure 6.4. Indirect client-server cooperation
for expanding the client-server model. (Source: Researchgate, h t t p s : / / w w w.
researchgate.net/profile/Tran-Vu-Pham-2/
publication/260388648/figure/fig1/AS:6694035
6.2.1. Agents and Indirect 02878734@1536609640717/Direct-and-indirect-
support-for-collaborations.ppm).
Client-Server Cooperation
Direct or indirect cooperation is possible Additionally, clients may implicitly
between a client and server. In the first request more than one service. As a result,
scenario, a client and a server exchange the client needs to contact a general server
requests and responses without the with just one request. Based on the data
involvement of a third party. In the included in the request, this invoked server
client-server model, indirect cooperation will work with other servers to create the
necessitates the request and provision of requested service. The invoked server
a service by two extra entities, referred to replies to the client after the involved
as agents. A similar extension is seen in servers have finished any necessary
Figure 6.4. processing. An agent with the right design
can carry out this coordination function.
These agents’ functions can range from The cooperation between the client and
a basic communication module that conceals the server is predicated on a single, clearly
communication network details to an entity
232 COMPUTER NETWORKS

defined interface, despite the complexity of such an agent. Additionally, the client
receives transparency, which lessens the application’s complexity.
A straightforward communication system that uses a direct, one-to-one message
protocol can facilitate cooperation between a client and several servers. Despite being
straightforward, this communication model performs poorly because it requires sending
separate messages to each of the servers involved. When a one-to-many communication
pattern is employed, the overall performance of a communication system supporting
message delivery in a client-server based distributed computing system can be significantly
increased. In this instance, the client process sends a single request—specified by a
single group name—to each server. Although it’s not necessary, multicasting at the
physical/data link layer does make this system better.

Q: How the browser interacts with the servers in while using client-server
model?

A: The Client-server model is a distributed application structure that partitions task or

workload between the providers of a resource or service, called servers, and service
requesters called clients. In the client-server architecture, when the client computer
sends a request for data to the server through the internet, the server accepts the
requested process and deliver the data packets requested back to the client. Clients
do not share any of their resources.
There are few steps to follow to interact with the servers a client.
• User enters the URL(Uniform Resource Locator) of the website or file. The
Browser then requests the DNS(DOMAIN NAME SYSTEM) Server.
• DNS Server lookup for the address of the WEB Server.
• DNS Server responds with the IP address of the WEB Server.
• Browser sends over an HTTP/HTTPS request to WEB Server’s IP (provided by
DNS server).
• Server sends over the necessary files of the website.
 DISTRIBUTED NETWORK SYSTEMS 233
• Browser then renders the files and the website is displayed. This rendering is
done with the help of DOM (Document Object Model) interpreter, CSS interpreter
and JS Engine collectively known as the JIT or (Just in Time) Compilers.

6.2.2. The Three-Tier Client-Server Architecture

The user interface (presentation), functional process logic (business
rules), computer data storage, and data access are all developed and Keyword
maintained as separate modules, usually on different platforms, in a Resource
three-tier architecture, a client-server software architecture pattern. management
software describes
The creator of it is John J. Donovan in Open Environment Corporation the digital tools
(OEC), a Cambridge, Massachusetts-based tools company he founded. used to allocate,
The three-tier architecture is meant to enable any of the three tiers assign, and
track employees,
to be upgraded or replaced independently in response to changes finances, and
in requirements or technology, in addition to the standard benefits equipment across
of modular software with well-defined interfaces. For instance, only projects.
the user interface code would be impacted by an operating system
change in the presentation tier. The functional process logic, which
can be comprised of one or more independent modules operating
on a workstation or application server, and the RDBMS, which is
a database server or mainframe that houses the computer data
storage logic, are typically used by the user interface, which runs
on a desktop PC or workstation and uses a standard graphical
user interface. If there are multiple tiers within the middle tier, the
overall architecture is referred to as an n-tier architecture.

Figure 6.5. Examples of three-tier configurations.

Source: https://www.setgetweb.com/p/WAS8/images/thtrcs.gif
234 COMPUTER NETWORKS

A well-known software application and is the final tier of the architecture. It

architecture known as “three-tier serves as a repository for the information
architecture” divides programs into three that has been processed and can be
logical and physical computing tiers: the retrieved as needed in the future. Database
application tier, which processes data, servers such as Oracle, MySQL, DB2, and
the data tier, which stores and manages so forth are included in it. Middle-tier I
application data, and the presentation facilitates the exchange of information
tier, which is the user interface. The main between the Presentation Tier and Data-
advantage of a three-tier architecture is Tier. E. Application Level.
that separate development teams can work
Table 6.1. The basics diffraction between Tiers
on each tier concurrently since each tier is
vs. layer
supported by its own infrastructure. and
can be modified or expanded as necessary
Tier Layer
without affecting the tiers below it. An
established software application design Tier refer to the Layer refers to the
pattern called “three-tier architecture” physical operation logical separation of an
divides an application into the following of components. application.
three logical and physical computing tiers: Tiers are physical Layers are logically
separated and separated but running
running on on the same servers or
Presentation Tier the different the machines.
machines are
It is the highest tier of the architecture and servers.
the user interface. Its function is to receive Scalability of an Scalability of an
client requests and provide information to application is application is medium.
them. It uses a web browser to communicate very high.
with other tiers and displays output on the
browser. When it comes to Web-based tiers, Common tiers Each layer focuses on
languages like HTML, CSS, and JavaScript in a multi-tier specific responsibilities,
are used in their development. architecture such as presentation,
include the business logic, and data
presentation tier access, within a single
Application Tier (user interface), tier.
application tier
As the information/request obtained through (business logic),
the presentation tier is processed in detail and data tier
here, it is the middle tier of the architecture (database).
referred to as the logic tier. The data storage
server is also a point of contact for it. After Source: https://www.geeksforgeeks.org/three-
handling and formatting the client’s request, tier-client-server-architecture-in-distributed-
it returns it to them. Languages like PHP, system/
Java, Python, and others are used in its
development.
Three-Tier Application In Web Development

Data Tier The three architectures and web applications

will have the same tiers; the only distinction
It is also referred to as the Database Tier is in their names.
 DISTRIBUTED NETWORK SYSTEMS 235
1. Web Server: A web server is a type of presentation tier in a three-tier architecture,
also known as a UI (User Interface) tier. Depending on the needs of the user—for
example, an e-commerce site where users can add items to their shopping cart,
enter payment information, or create an account—this content can be static or
dynamic.
2. Application Server: The application tier is a middle tier that houses all of the
logic needed by the company to process user inputs and other data. It will
carry out logical operations such as updating a customer’s profile with new
information or querying the inventory database to retrieve product availability.
3. Database server: The back-end tier of a web application houses all of the data,
so databases like MySQL, Oracle, DB2, or PostgreSQL are typically found here.

Figure 6.6. An example implementation for creating a three-tier web application.

Source: https://www.gstatic.com/pantheon/images/solutions/three_tier_web_app_v6.svg

Benefits of a tree-tier app architecture

A three-tier architecture offers enhanced availability, performance, and horizontal
scalability. Three tiers allow for the simultaneous development of each component
by distinct programming teams using different languages from those of the other tier
developers. An enterprise or software packager can more easily evolve an application
as new needs and opportunities arise thanks to the 3-tier model, which allows tier
programming to be changed or relocated without affecting other tiers. Critical components
or applications that are already in place can be contained within the new tier they are
236 COMPUTER NETWORKS

a part of, either permanently or temporarily. N-tier programs are another name for 3-tier
application programs. The letter n represents several tiers in this context.
• Logical separation is maintained between Presentation Tier, Application Tier,
and Database Tier.
• Enhancement of Performance as the task is divided on multiple machines in
distributed machines and moreover, each tier is independent of other tiers.
• Increasing demand for adding more servers can also be handled in the architecture
as tiers can be scaled independently.
• Developers are independent to update the technology of one tier as it would
not impact the other tiers.
• Reliability is improved with the independence of the tiers as issues of one tier
would not affect the other ones.
• Programmers can easily maintain the database, presentation code, and business/
application logic separately. If any change is required in business/application
logic then it does not impact the presentation code and codebase.
• Load is balanced as the presentation tier task is separated from the server of
the data tier.
• Security is improved as the client cannot communicate directly with Database
Tier. Moreover, the data is validated at Application Tier before passing to
Database Tier.
• The integrity of data is maintained.
• Provision of deployment to a variety of databases rather than restraining yourself
to one particular technology.

Did you know?

Prior to the 1980s, computing was typically centralized on a single low-cost desktop computer. But today, computing
resources (computers or servers) are typically physically distributed in many places, at which distributed networking
excels.

Did you get it?

1. What role does the client typically play in the client-server model when it comes to user interface
and presentation processing?
2. How does the server contribute to data access processing within the client-server model?
6.3. SERVICE
DISCOVERY
→Learning Objectives

• Understand the concept of hardwiring computer addresses and its significance

in networking.
• Explain the broadcast approach in networking and its use for communication
between devices.
• Describe the name server approach and its role in resolving domain names to
IP addresses.
• Analyze the broker-based location lookup method and its application in distributed
systems.
• Explore the concept of client-server interoperability and its importance in
heterogeneous network environments.
• Define client/server computing and identify its key components and characteristics.

To request a desired service, a client needs to know if there is a server that can
provide it, what those characteristics are, where it is located, and its name. This is the
service discovery problem. It is not necessary to determine whether a desired server
exists in a simple distributed computing system with a small number of servers. All of
the servers that are available are known in advance. This suggests that the scope of
service discovery is limited to identifying the server that offers the required service.
However, in a sizable distributed computing system that is a federation of several
dispersed computing systems, with numerous service providers that dynamically offer
and withdraw these services, it is necessary to ascertain whether a suitable service
(e.g., a high-quality, extremely quick color printer) is available at a specific time, along
with its name and location.
One of the most crucial aspects of a distributed computing system is transparency.
It is important to note that in a distributed computing system managed by a distributed
operating system, a client should only be aware of the name of a server or an agent
acting on the server’s behalf. In contrast, a client in a distributed computing system
that is controlled by a collection of centralized operating systems and their extensions
needs to be aware of its name and location in order to access remote resources and

DISTRIBUTED NETWORK SYSTEMS 237

238 COMPUTER NETWORKS

services. The absence of transparency is send a request, a client needs to know

the cause. the name of the destination, specifically
the server. This is insufficient, though,
Service discovery is achieved through since the server’s computer’s operating
the following modes: system needs to be aware of its address.
• Server computer address is To achieve this, every computer connected
hardwired into client code; to a network will receive a unique locate
request from the client’s operating system
• Broadcast is used to locate servers;
that includes the server’s name. When an
• Name server is used to locate operating system discovers the name of the
services; and server in the list of processes it is running,
• Brokers are used to locate servers. indicating that the server is running on
its machine, it responds with a “here I
am” message that includes the address
6.3.1. Hardwiring Computer (location). After receiving the response, the
Address operating system of the client can cache—or
All that is needed for this method is the save—the server’s computer address for use
server’s location, which takes the form of in subsequent communications. Although
its computer address. It is only useful in all computers on a network are involved in
extremely small, straightforward systems, processing the locate request, this method
though, where the destination computer is transparent but has a high broadcast
is only hosting one server process. An overhead.
operating system can therefore deliver an
incoming request to the right place.
A more sophisticated naming scheme
forms the basis of an alternative variation
of this methodology, in which requests are
routed through processes as opposed to
computers. Each process in this instance
is identified by a pair <computer_address,
process_name>. In addition to the server’s
name, a client receives the server computer’s
address. Since the user is aware of the
server’s location, this solution is not location
transparent. When a server needs to be
moved to a different computer and a pair
of computers, the lack of transparency can
cause issues. <computer_address, process_
name> is ingrained in the client code. Figure 6.7. Service discovery -- broadcast
approach (Source: Middleware.io, https://
middleware.io/wp-content/uploads/2021/09/
6.3.2. Broadcast Approach Client-Side-Service-Discovery-768x713.jpg).
This method assigns a unique name to each The cooperation between clients,
process (e.g., for this purpose, an extremely servers, and operating systems supporting
lengthy identifier can be employed). To them in a distributed computing system
 DISTRIBUTED NETWORK SYSTEMS 239
using the broadcast approach to locate servers is illustrated in
Figure 6.7.

6.3.3. Name Server Approach

This method reduces the broadcast overhead but is quite similar to
the broadcast-based method. The client’s operating system sends a
“where is” request to a name server—a specialized system server—
asking for the address of the machine running the desired server
to locate the desired server. All operating systems are aware of
the name server’s location (computer address) and name. The name
server returns the requested server’s address in a response. Upon
receiving the response, the client’s operating system can store the
server’s computer address for future communications.
When comparing this approach to the broadcast-based method,
it is transparent and much more effective. However, due to the
centralized nature of the name server, it may become a bottleneck
and impact the overall performance of a distributed computing
system. Additionally, this method lacks reliability; if a name server
computer crashes, a distributed computing system cannot operate.
Collaboration among clients, servers, and the supporting operating
systems is essential in a distributed computing system utilizing
the server.

6.3.4. Broker-Based Location Lookup

To access a desired service in a distributed computing system
that uses any of the aforementioned server location approaches, a
client needs to be aware of all the servers that are present in the
system, their names, and, in the case of non-transparent systems,
the locations of the servers (addresses). There may be a lot of
servers in a large distributed computing system. In addition, distinct
attributes pertaining to the services that they offer can distinguish Keyword
servers of the same kind (e.g., a black and white printer and a
Service
color printer are two types of laser printers). Servers can also be Management
dynamically revoked and offered by certain users. The identities, focuses on
characteristics, and constantly fluctuating availability of each of providing value
to the customer
these servers are not visible to the user. A server must exist that and also on
could assist users in resolving these issues. A broker is the term the customer
for this server. Therefore, a broker is a server that: relationship.

• allows a client to identify available servers which can

be characterized by a set of attributes that describe the
properties of a desired service;
• mediates cooperation between clients and servers;
240 COMPUTER NETWORKS

• allows service providers to register The forwarding broker possesses

the services they support by advantages and disadvantages of a system
providing their names, locations, with a centralized server. This means that
and features in the form of all requests to servers and their responses
attributes; are going through this broker.
• advertises registered services and
makes them available to clients; Direct Broker
and
Cooperation between a client and a server
• withdraws services dynamically.
mediated by this broker is as follows:
A name server approach’s server
location lookup is extremely similar to that • Step 1: The broker receives from
of a broker-based approach. A broker and a client a service inquiry in the
a name server, on the other hand, differ form of a set of attributes that
conceptually in important ways. A broker characterize a desired service;
relieves clients of the burden of memorizing • Step 2: If a matching server is
the path names or ASCII names of every available, the broker sends back
server (and eventually their locations), while a name and a server computer
also enabling clients to identify server address to the client; otherwise,
attributes and availability. As a result, a it sends a failure response;
broker is a server that combines naming
• Step 3: The client sends the server
and service management functions. There
operation request to the server;
exist two fundamental categories of brokers,
representing distinct modes of collaboration • Step 4: The server sends back a
between clients and servers: response to the client.
Despite the fact that the direct
6.3.4.1. Forwarding Broker broker also possesses advantages and
Cooperation between a client and a server disadvantages of a system with a centralized
mediated by this broker is as follows: server, its performance is better than that of
the forwarding broker because only service
• Step 1: The broker receives from inquiry messages are sent to this broker.
a client a service inquiry in the
form of a set of attributes that
characterize a desired service, and 6.3.5. Client-Server
a server operation request; Interoperability
• Step 2: If a matching server is Because software writing is expensive,
available, the broker sends the server reusability is a crucial concern for
server operation request to that both users and software manufacturers.
found server; otherwise, it sends a In a homogeneous environment, this
failure response to the client; problem might be easily fixed because
• Step 3: the server sends back a software interfaces can be used to make
response to the broker; client access mechanisms compatible, with
• Step 4: the broker passes the static compatibility determined by types
response to the client. and dynamic compatibility by protocols.
Because they are not entirely compatible,
cooperation between heterogeneous
 DISTRIBUTED NETWORK SYSTEMS 241
clients and servers is far more challenging.
Making them interoperable is therefore 6.3.6. Client/Server Computing
the problem. The capacity of two or more The idea of dividing up related processing
software components to work together across multiple resources has gained
despite variations in language, interface, popularity as networks of computing
and execution platform is known as resources have become more commonplace.
interoperability. Many techniques have developed over time
to make this distribution easier. Client/
Client-server interoperability has two server computing is one of the most widely
components: interoperation mechanisms and used distributed models. The modular
an interoperation unit. A procedure is the programming model is expanded upon
fundamental building block of cooperation. by the client/server model. A program’s
However, software components might need design can be divided into discrete modules
higher-granularity units of interaction. that can be independently programmed
Additionally, it might be necessary to and tested through the use of modular
preserve the functional and temporal programming. A main module and one or
properties. more auxiliary modules make up a modular
There are two major mechanisms for program. Clients and servers make up a
interoperation: client/server model, just like in a modular
programming model. Typically, a computer
• Interface Standardization: The network connects the various computers
objective of this mechanism is to that the clients and servers use to operate.
map client and server interfaces The called component is the server, and the
to a common representation. The calling component is the client.
advantages of this mechanism
are: (i) it separates communication Through the network, a client application
models of clients from those communicates with a server to ask it to do
of servers, and (ii) it provides a particular task. Input-output devices, local
scalability, since it only requires m disks, and other peripherals are managed
+ n mappings, where m and n are by the client. Requests from clients that
the number of clients and servers, are sent over the network are received by
respectively. The disadvantage of the server program. After receiving those
this mechanism is that it is closed. requests, servers take certain actions. Only
the results are sent back to the client after
• Interface Bridging: The objective of the majority of the data has been processed
this mechanism is to provide a two- on the server. As a result, there is less
way mapping between a client and network traffic between the client computer
a server. The advantages of this and the server. Network performance is
mechanism are: (i) openness, and further enhanced. In addition to managing
(ii) flexibility — it can be tailored to information distribution, the server optimizes
the requirements of a given client resource usage.
and server pair. However, this
mechanism does not scale as well Whether a client communicates with the
as the interface standardization server directly or whether an intermediary
mechanism, as it requires m * n process is introduced between the client and
mappings. the server is a crucial design consideration
for large client/server systems. Web
242 COMPUTER NETWORKS

applications typically use N-tier architecture to forward requests

to additional enterprise services. Small environments are where
the two-tier architecture is most commonly utilized because it is
simpler to implement. Three-tier architecture is more scalable than
two-tier architecture, though.
An intermediary process links the clients and servers in the
three-tier architecture. For improved speed and scalability, the
middleman can gather frequently used server data. A client PC, an
application server, and a database server are typically the three
main parts of a database-based 3-tier client/server architecture. The
middle tier server that manages the business application is called
the application server. The middle tier server retrieves the data from
the database server and forwards it to the client. Creating a three-
tier client/server application requires middleware. An Application
Program Interface (API) is provided by database-oriented middleware
to access a database. The popular Java Database Connectivity
(JDBC) classes can be used to help an applet or servlet access
multiple databases without taking into account the database’s
internal features.
Keyword
The issue of authentication must also be addressed by servers
Application
programming for security reasons. Sensitive information kept on a server by an
interface (API) unauthorized client could be accessed in a networked environment.
is a way for two Cryptographic techniques like public key encryption and special
or more computer
programs or authentication servers are used to provide client authentication.
components to Replicating vital servers can occasionally be done to attain fault
communicate with tolerance and high availability. Clients can still access the other
each other. It is a
type of software replicas hosted on separate servers in the event that one fails.
interface, offering
a service to other It is simpler to change or replace any tier in a three-tier
pieces of software. architecture without impacting the other tiers. Improved load
balancing is achieved by separating the functionality of the database
and application. Furthermore, without negatively affecting the
clients, the server tiers can implement the required security policies.
The “proxy server model,” a three-tier client/server architecture
(Figure 6.7), is frequently used to enhance Internet retrieval
performance. In order to allow requests to run concurrently, the
intermediary process known as the proxy server divides up client
requests among multiple servers. A client establishes a connection
with the proxy server and requests a service, like a web page
hosted on a web server. The request is evaluated by the proxy
server according to its filtering policy. It might, for instance, filter
traffic based on a protocol or IP address. The proxy presents the
resource by connecting to the relevant server and requesting the
necessary service for the client if the filter authenticates the request.
 DISTRIBUTED NETWORK SYSTEMS 243
Sometimes a proxy server will fulfill the request without getting in touch with the
designated web server. This is made possible by storing frequently visited pages in
the proxy’s cache. The proxy can serve frequently accessed files back to a requesting
browser without having to visit the external site each time by maintaining local copies
of those files; this significantly enhances the end user’s experience with performance. A
“proxy-cache server” is the term used to describe a proxy server that has the capacity
to cache data. A stand-in is occasionally.

Did you get it?

1. How does the broker-based location lookup method function, and what are its applications in
distributed systems?
2. What is the definition of client/server computing, and what are the key components and
characteristics associated with this model?
6.4. CHARACTERISTICS OF
A DISTRIBUTED SYSTEM
→Learning Objectives

• Understand the concept of fault tolerance and its importance in ensuring system
reliability and resilience.
• Explain scalability in the context of computing systems and its implications for
accommodating growing workloads or user demands.
• Define predictable performance and discuss strategies for achieving consistent
and reliable system response times.
• Explore the concept of openness in software and systems, including the benefits
of interoperability and accessibility.
• Analyze security considerations in computing environments, including threats,
vulnerabilities, and strategies for safeguarding data and resources.
• Discuss transparency in system design and operation, including mechanisms for
providing visibility and insight into system behaviors and processes.

Distributed systems are characterized by several key features that distinguish them
from centralized or standalone systems. These characteristics define the architecture,
behavior, and capabilities of distributed systems, enabling them to achieve scalability,
fault tolerance, and efficient resource utilization. One fundamental characteristic of
distributed systems is decentralization, where multiple autonomous nodes collaborate
to perform tasks or provide services without relying on a central coordinator. This
decentralization fosters resilience and flexibility, as nodes can operate independently
and adapt to changes in the network topology or workload distribution without affecting
the system’s overall functionality.
Another important characteristic of distributed systems is concurrency, which refers
to the ability to perform multiple tasks or processes concurrently across different nodes
in the network. Concurrency enables distributed systems to handle multiple requests
simultaneously, improving throughput and responsiveness. However, managing concurrency
requires mechanisms for synchronization and coordination to ensure consistency and
integrity of shared resources or data accessed by multiple concurrent processes.
Scalability is a critical characteristic of distributed systems, allowing them to
accommodate growing workloads or user demands by adding more nodes to the network.
Scalability can be achieved through horizontal scaling, where additional nodes are added to
distribute the workload, or vertical scaling, where existing nodes are upgraded to handle

244 COMPUTER NETWORKS

 DISTRIBUTED NETWORK SYSTEMS 245
increased capacity. Distributed systems to interact with distributed systems
must be designed to scale efficiently, seamlessly without needing to be aware of
ensuring that adding or removing nodes the underlying complexities of the network
does not introduce bottlenecks or degrade architecture.
performance.
A distributed system must possess the
Fault tolerance is another key following characteristics to deliver utmost
characteristic of distributed systems, performance for the users:
enabling them to continue operating
even in the presence of node failures,
network partitions, or other types of 6.4.1. Fault-Tolerant
faults. Distributed systems achieve fault Distributed systems are composed
tolerance through redundancy, replication, of numerous hardware and software
and distributed consensus mechanisms components, all of which will eventually
that enable nodes to detect and recover fail. These types of component failures can
from failures autonomously. By replicating result in service disruptions. Therefore,
data or services across multiple nodes and systems should be able to recover from
implementing fault detection and recovery component failures without errors. Fault
mechanisms, distributed systems can tolerance is designed to prevent system
maintain availability and reliability in the failures even in the presence of faults to
face of failures. ensure continuous service. If a system can
hide errors, it is considered fault tolerant.
Heterogeneity is a characteristic
The objective of any fault-tolerant system is
commonly found in distributed systems,
to enhance its dependability and availability.
where nodes may vary in terms of
The system’s ability to operate until a
hardware, operating systems, programming
certain point is known as its reliability.
languages, or communication protocols.
Dependable systems prevent data loss
Heterogeneity introduces challenges related
even in the event of component failures.
to interoperability, data consistency, and
The time a system is available for use is
resource management, requiring distributed
referred to as its availability. Redundancy
systems to support diverse environments
is typically employed to achieve fault
and integrate seamlessly with different
tolerance, involving components that are not
technologies.
essential for the system’s basic operation.
Finally, distributed systems exhibit There are three types of redundancy:
transparency, which refers to the extent time, software, and hardware. Hardware
to which the distribution of resources redundancy is achieved by adding extra
and processes is hidden from users and hardware components to take over the
applications. Transparency encompasses functions of failed components.
aspects such as location transparency,
where the physical location of resources
is abstracted from users, and access
6.4.2. Scalable
transparency, where users can access When a distributed system is scaled to
distributed resources as if they were a larger size, some parts of the system
centralized. Achieving transparency still function as intended. Three factors
simplifies the development and deployment determine a system’s scale: the quantity
of distributed applications, enabling users of users and other entities it contains, the
246 COMPUTER NETWORKS

separation between its farthest nodes, the data among several servers, the size
and the number of organizations that have of the database that each server needs to
administrative control over certain areas maintain is decreased, which cuts down on
of the system. The three dimensions of the amount of time required for database
scale have a variety of effects on distributed searches. Additionally, distribution divides
systems. A few of the components that the load among the servers, limiting the
are impacted are naming, authorization, amount of requests that each can handle.
communication, remote resource utilization, The load on servers can be efficiently
authentication to confirm an individual’s managed if requests are routed to servers
identity, and user observation mechanisms. according to their power. Assigning data
To control scale, three strategies are used: to servers near their most frequently used
caching, replication, and distribution. location can help minimize network traffic.
If cached copies are accessible from lower-
Resources are replicated multiple level servers in a tree-structured system,
times during replication. Its application to the higher levels can be skipped.
file services, naming, and authentication
lightens the burden on individual servers Another crucial method for creating
and raises the overall availability and scalable systems is caching. The strain
dependability of the services. The locations on servers and the network is reduced
of the replicas and the methods used to by caching. Accessing data that has been
maintain consistency between them are the cached is quicker than making a new
two key aspects of replication. The reason request. Replication and caching differ in
for replicating a resource determines where that the latter is a temporary storage of
replicas should be placed in a distributed data. When consistency cannot be ensured,
system. A service’s replicas are dispersed consistency is preserved by nullifying cached
throughout the system if it is being data rather than propagating updates on
replicated in order to minimize network it. As the client typically handles caching,
latency when the service is accessed. there are fewer repeated requests made
Replicas may be positioned close to one to the network services. On the servers
another if most users are local and if the running those services, caching may also
service is being replicated to increase take place. It is quicker to read a file from
availability or distribute the load across the file server’s memory cached copy than
several servers. Any modifications made to from the client’s local disk.
the object should be apparent to all users
of the system. As an illustration, the system
updates any replica it finds, and that replica 6.4.3. Predictable Performance
then distributes the updates to the others as A range of performance indicators, including
they become available. Timestamps, or the system utilization, network capacity,
date and time the update was generated, throughput (the speed at which data is
are used to distinguish between copies of sent or received over a network), response
an inconsistent update if they are received time (the amount of time that passes
by various replicas in different orders. between the end of a request or inquiry
on a computer system and the start of a
A distributed service’s data can be
response), etc., are used in order to evaluate
expanded across multiple servers through
the performance. The capacity to deliver
distribution, another method for controlling
desired responsiveness on time is known
scale in distributed systems. By spreading
as predictable performance.
 DISTRIBUTED NETWORK SYSTEMS 247
6.4.4. Openness
The quality that guarantees a subsystem is always available for
communication with other systems is called “openness.” Software
systems known as web services are made to facilitate networked,
interoperable machine-to-machine communication. These protocols
enable scalability and extension of distributed systems. A scalable, Remember
open system is preferable to one that is totally closed and A distributed
independent. The openness of a distributed system is achieved networking
system is
without relying on the heterogeneity of the underlying environment,
composed of
including hardware and software platforms. As a result, each service processes,
in the system is equally available to all clients, whether they threads, agents,
are local or remote. New service implementation, installation, and and distributed
debugging shouldn’t be too difficult in a system with an openness objects. Merely
distributed
characteristic. physical
components
are not enough
6.4.5. Security to suffice as
a distributed
By enforcing the appropriate security measures, distributed systems network; typically
should enable communication between users, programs, and distributed
resources on various computers. The primary goals of the security networking
features are to ensure availability, integrity, and confidentiality. uses concurrent
program
Privacy, or confidentiality, is the defense against revelation to
execution.
unauthorized parties. Confidentiality violations can range from
upsetting to disastrous. Integrity shields against corruption and
modification. Availability maintains the resource’s usability. The
integrity of databases and other resources is frequently compromised
by hacking incidents. The term “denial of service” refers to an attack
that targets availability. Nonrepudiation and access control are two
more crucial security issues. Ensuring that users can only access
resources and services to which they are legally entitled is made
easier by maintaining access control. Additionally, it makes sure
that users don’t lose access to resources that they have a right to.
Protection against denial by one of the parties to a communication
is offered by nonrepudiation. The security measures implemented
ought to ensure that various system users are utilizing resources
in an appropriate manner.

6.4.6. Transparency
Rather than being seen as a group of collaborating parts, distributed
systems should be viewed as a whole by users and application
developers. The locations of computer systems that are involved in
concurrent operations, data replication, failures, system recovery,
and resource discovery from multiple sites. are concealed from
users. Transparency allows the system to appear and function like
248 COMPUTER NETWORKS

a typical centralized system, concealing from its users the fact that it is distributed. In
a distributed system, the transparency can be used in various ways (Figure 6.8).

Figure 6.8. Transparency in Distributed Systems (Source: Geeksforgeeks, https://media.

geeksforgeeks.org/wp-content/uploads/20220527122938/Dimensionsoftransparency.jpg).

• Access transparency facilitates the users of a distributed system to access local

and remote resources using identical operations (e.g., navigation on the web).
• Location transparency describes names used to identify network resources (e.g.,
IP address) independently of both the user’s location and the resource location.
In other words, location transparency enables a user to access resources from
anywhere on the network without knowing where the resource is located. A
file could be on the user’s own PC or thousands of miles away on other servers.
• Concurrency transparency enables several processes to operate concurrently
using shared information objects without interference between them (e.g.,
Automatic Teller Machine network). The users will not notice the existence of
other users in the system, even if they access the same resources.
• Replication transparency enables the system to make additional copies of files
and other resources for the purpose of performance and/or reliability, without
the users noticing. If a resource is replicated among several locations, it should
appear to the user as a single resource (e.g., Mirroring – Mirror sites are usually
used to offer multiple sources of the same information as a way of providing
reliable access to large downloads).
• Failure transparency enables applications to complete their tasks despite failures
occurring in certain components of the system. For example, if a server fails,
but users are automatically redirected to another server and the user never
notices the failure, the system is said to exhibit high failure transparency.
Failure transparency is one of the most challenging types of transparency to
achieve since it is difficult to determine whether a server has actually failed or
is simply responding very slowly. Moreover, achieving full failure transparency in
a distributed system is generally unfeasible due to the unreliability of networks.
• Migration transparency allows resources to move from one location to another
 DISTRIBUTED NETWORK SYSTEMS 249
without changing their names (e.g., Web Pages). Users should not be aware
of whether a resource or computing entity can move to a different physical or
logical location.
• Performance transparency ensures that load variation does not result in
performance degradation. This can be achieved through automatic reconfiguration
in response to changes in the load (e.g., load distribution).
• Scalability transparency enables the system to remain efficient even with a
significant increase in the number of users and connected resources (e.g., World-
Wide-Web, distributed database).

Did you get it?

1. What does scalability mean in the context of computing systems, and how does it impact
the ability to handle increasing workloads or user demands?
2. How can you analyze security considerations in computing environments, including identifying
potential threats and vulnerabilities, and implementing strategies to protect data and resources?
250 COMPUTER NETWORKS

FOCUS ON CAREERS
WIRELESS NETWORK ENGINEER
A Wireless Network Engineer job description includes installing, configuring, and
maintaining wireless network equipment, network management, and security, including
802.11 b/g/n/ac standards and industry best practices for implementing high-density WIFI
solutions. Moreover, to put it concisely, a Wireless Network Engineer needs to assess,
plan, and develop several operational capabilities for wireless telecommunications. As
more companies continue to embrace Wireless LAN, the demand for wireless network
engineers has grown manifold in the recent past. The proliferation of mobile applications,
which require testing in a wireless environment, is also one of the other reasons why
enterprises need their services more. Research indicates that 60% of enterprise-level
managers report utilizing freelancers in their workforce. But with the emergence of
Software Defined WAN (SD-WAN) DevOps, the advent of 5G, and virtualization, it is
necessary for Wireless Network Engineers to be open to reskill themselves, as it could
likely bring about changes that might change the way they operate. It is, however,
averred by experts that despite all the changes that may take place in the networking
landscape, the demand for Wireless Network Engineers will continue to rise.

Wireless Network Engineer Job Description

To become a freelance Wireless Network Engineer, aspirants need to be well-versed in
wireless equipment, wireless LAN (WLAN) standards, design, and protocols. Wireless
Network Engineer job descriptions will require excellent analytical and problem-solving
skills. They should also have good communication skills as they would need to work
alongside vendors, network technicians, and customers.
Wireless Network Engineers need to be available round-the-clock to support mission-
critical applications. They should have strong customer service skills.
The responsibilities and duties of a Wireless Network Engineer include the following:
• Devise, plan, deploy, and improve wireless networks from the beginning to
implementation by collaborating with vendors, managers, and network engineers.
• Understand client requirements to be able to cater to their appropriate needs.
• Manage firewalls, such as Palo Alto, Juniper, or Cisco ASA.
• Handhold other in-house engineers to train them on wireless technologies,
besides guiding other non-technical people.
• Use tools to evaluate to test and tweak wireless products, such as routers,
switches, hubs, bridges, virtual private networks (VPNs), and network amplifiers,
among others.
• Design and validate the performance, quality, and reliability of the RF link.
 DISTRIBUTED NETWORK SYSTEMS 251

• Design and implement WLANs and other wireless networks.

• Write manuals and document current network procedures.
• Have extensive knowledge of routing protocols (OSPF, EIGRP, and BGP).Make
use of enterprise monitoring tools (SolarWinds and Splunk).
• Optimize network performance by supervising performance, addressing network
problems and breakdowns, and partnering with network engineers to optimize
the network.
• Develop and implement policies and classify and oversee access to protect
network systems.
• Ensure that all equipment, including servers and other network products, is
well-connected.
• Devise and support radio frequencies (RF) link performance, reliability, and quality.
• Enable data and system protection by developing and maintaining mechanisms
for backing up and retrieval.
• Collaborate with different teams to ensure the optimized performance of VoIP
and other wireless telecommunication devices.
• Work with the security team to evaluate threats, troubleshoot issues, and comply
with appropriate security configuration standards of their organizations.
• Design and deploy changes to the configurations of clients as per the applicable
change management process.
• Proactively handle all network security solutions.
• Utilize testing and diagnostic tools to assess and modify equipment.

Wireless Network Engineers Job Prospects

As more companies continue to embrace Wireless LAN, the demand for wireless network
engineers has recently grown. The proliferation of mobile applications, which require
testing, is also one of the reasons why enterprises need their services more.
With the emergence of Software Defined Networking (SDN), DevOps, 5G, and virtualization,
it is necessary for an engineer to be open to reskilling themselves, as it could likely
bring about changes that will alter the way networks operate.
It is, however, asserted by experts that despite all the changes that may take place in
the networking landscape, the demand for Wireless Network Engineers will continue
to grow.

Wireless Network Engineer Training & Educational Qualifications

Most employers would look for Wireless Network Engineers who have undergone wireless
network engineer training to earn a bachelor’s degree or an equivalent in mathematics,
engineering, or computer science. They should have a certification in Cisco Certified
252 COMPUTER NETWORKS

Network Professional (CCNP) Wireless or Certified Wireless Network Professional (CWNP)

or Aruba Certified Mobility Associate (ACMA).
Since it is a skill set that requires experience, a minimum experience of five years
in WAN/LAN engineering is mandatory for an aspirant.
Through wireless network engineer training, they should become adept with wireless
technologies, such as Wi-Fi, WiMax, and WAP. Wireless Network Engineer Training should
have a thorough knowledge of 802.11n and 802.11a. They should be able to understand
scripting languages, such as Bash, Perl, or Python.
Engineers should have extensive knowledge of both wired and wireless networking.

Wireless Network Engineer Salary

According to Indeed, the average earnings of a Wireless Network Engineer (CCNP
certified) range in a senior position from about $87,772 per year to $106,566 per year in
the United States. While their average annual earnings are $83,998 per year, freelancers
earn $24.57 per hour, says Payscale. There are numerous additional factors that make
up the salary of a Wireless Network Engineer.
 DISTRIBUTED NETWORK SYSTEMS 253
SUMMARY
• A group of linked computers is referred to as a computer network. Nodes designate
computers connected to a network. Computers can be linked wirelessly using
radio waves or through cabling, typically an Ethernet cable..
• Wide Area Networks (WANs) link computers and local networks to larger
networks across larger geographical regions, such as continents.
• An assembly of computing hardware, including computer systems, connected via
communication channels to enable resource sharing and communication among
users, forms a computer network.
• The fundamental idea behind OSI is that communication between two endpoints
in a network can be separated into seven different groups, or layers, of related
functions. Every user or program that communicates uses a device capable of
providing those seven levels of functionality.
• Routing protocols are designed for internet routers to use.
• In a ring topology, each device has a unique connection to the two devices on
its sides..
• Every device in a star topology has a dedicated point-to-point link that connects
it to the hub, or central controller.
• Hostname is a special device name that is assigned to each device in the
network. In the Administrator Mode command prompt, type “hostname” and
hit “Enter” to see your machine’s hostname.

MULTIPLE CHOICE QUESTIONS

1. In distributed system each processor has its own
a. local memory
b. clock
c. both local memory and clock
d. none of the mentioned
2. If one site fails in distributed system
a. the remaining sites can continue operating
b. all the sites will stop working
c. directly connected sites will stop working
d. none of the mentioned
3. A dynamic connection that grows into dynamic networks of networks, is called
a. Cyber cycle
b. Internet of things
c. Cyber-physical system
d. Multithreading
254 COMPUTER NETWORKS

4. Parallel computing is also known as

a. Parallel computation
b. Parallel processing
c. Parallel distribution
d. Parallel development
5. Connections that grows exponentially into a new dynamic network of networks, is
known as
a. Dynamic
b. Static
c. Transparent
d. Opaque

REVIEW QUESTIONS
1. Define the client-server model in a distributed computing system.
2. Explain the cooperation between clients and servers.
3. What are the extensions to the client-server model?
4. Discuss the three-tier client-server architecture.
5. What is broadcast approach?

Answers of Multiple Choice Questions

1. (c) 2. (a) 3. (b) 4. (b) 5. (a)

REFERENCES
1. Robert, R., Chris, A., & Jay, L., (2003). A solver for the network testbed mapping
problem. ACM SIGCOMM Computer Communications Review, 32(2), 65–81.
2. Jonathon, D., Robert, R., John, B., & Jay, L., (2006). Automatic IP Address Assignment
on Network Topologies. Technical Report FTN-2006-02.
3. David, S. A., Leigh, S., Mike, H., Tim, S., & Jay, L., (2006). Automatic online validation
of network configuration in the Emulab network testbed. Third IEEE International
Conference on Autonomic Computing (ICAC 2006).
4. Ang, B. S., Chiou, D., Rosenband, D. L., Ehrlich, M., Rudolph, L., & Arvind, (1998).
StarT-voyager: A flexible platform for exploring scalable SMP issues. In: Proceedings
of SC98: High Performance Networking and Computing. Orlando, Florida.
5. Arpaci-Dusseau, C., (1998). Implicit Coscheduling: Coordinated Scheduling with
Implicit Information in Distributed Systems. PhD thesis, University of California at
Berkeley.
 DISTRIBUTED NETWORK SYSTEMS 255
6. Adem, E. G., David, B., Emin, G. S., & Robbert, V. R., (2015). Configuring Distributed
Computations Using Response Surfaces. Middleware, Vancouver, Canada.
7. Alysson, N. B., Eduardo, A., Miguel, C., & Joni, S. F., (2008). DepSpace: A Byzantine
Fault-Tolerant Coordination Service. Proceedings of the 3rd ACM SIGOPS/EuroSys
European Systems Conference – EuroSys’08.
8. Jonathon, D., Robert, R., Junxing, Z., Daniel, G., Sneha, K., & Jay, L., (2006). Flexlab:
A realistic, controlled, and friendly environment for evaluating networked systems.
Fifth Workshop on Hot Topics in Networks (HotNets-V).
9. Shashi, G., Rob, R., & Jay, L., (2005). Integrated network experimentation using
simulation and emulation. First International Conference on Testbeds and Research
Infrastructures for the Development of Networks and Communities (Tridentcom 2005).
10. Shashikiran, B. G., (2005). Issues in Integrated Network Experimentation Using
Simulation and Emulation. Thesis, University of Utah.
11. David, J., Tim, S., Russ, F., Daniel, M. F., Leigh, S., Robert, R., & Jay, L., (2006).
Mobile Emulab: A Robotic Wireless and Sensor Network Testbed. IEEE INFOCOM
2006.
12. Arora, N. S., Blumofe, R. D., & Plaxton, C. G., (1998). Thread scheduling for
multiprogrammed multiprocessors. In: Proceedings of the 10 th Symposium on Parallel
Algorithms and Architectures (pp. 119–129). Puerto Vallarta, Mexico.
13. Jonathon, D., Robert, R., John, B., & Jay, L., (2005). Optimizing IP Address Assignment
on Network Topologies. Technical Report FTN-2005-04, University of Utah.
14. Bruno, V., Nuno, N., Henrique, M., & Paulo, V., (2010). Randomized consensus in
wireless environments: A case where more is better. Third International Conference
on Dependability. Venice, Italy.
15. Nitin, G., Alan, J. D., & Johannes, E. G., (2008). SEMMO: A scalable engine for
massively multiplayer online games. Talk. International Conference on Management
of Data. Vancouver, Canada.
16. Vincent, G., Ymir, V., Ken, B., Anne-Marie, K., & Robbert, V. R., (2007). Sliver, a
Fast Distributed Slicing Algorithm. Technical Report.
17. David, J., Daniel, G., & Jay, L., (2008). Towards a high quality path-oriented
network measurement and storage system. Proc. of the Ninth Passive and Active
Measurement Conference (PAM 2008).
18. Michael, P. K., Priya, N., Kevin, A., & Jay, L., (2006). Towards finger pointing in
the Emulab dynamic distributed system. Third USENIX Workshop on Real, Large
Distributed Systems (WORLDS 2006).
19. David, J., Tim, S., Russ, F., Dan, F., Rob, R., & Jay, L., (2005). TrueMobile: A Mobile
Robotic Wireless and Sensor Network Testbed. Technical Report FTN-2005-02,
University of Utah.
CHAPTER
Wireless
7 Network Security

LEARNING OBJECTIVES

After studying this chapter, you will be able to:

• Define the synopsis of wireless technology

• Examine the development of wireless LANS
• Learn about the wireless personal area networks
258 COMPUTER NETWORKS

INTRODUCTORY EXAMPLE

WIRELESS NETWORK SECURITY IN AN ORGANIZATION

Wireless network security in an organization is a critical aspect of ensuring the
confidentiality, integrity, and availability of sensitive information and resources. It
encompasses various measures, protocols, and best practices designed to protect the
organization’s wireless infrastructure from unauthorized access, data breaches, and
cyber threats.
To begin with, implementing robust authentication and encryption mechanisms is
essential. Strong authentication protocols like WPA2-Enterprise or WPA3, coupled with
complex pre-shared keys (PSKs) for smaller deployments, help verify the identity of users
and devices before granting access to the network. Encryption protocols such as AES
(Advanced Encryption Standard) ensure that data transmitted over the wireless network
remains secure and cannot be intercepted or tampered with by unauthorized parties.
Furthermore, controlling access to the wireless network is vital for minimizing
security risks. This can be achieved through techniques such as MAC address filtering,
where only devices with pre-approved MAC addresses are allowed to connect, and
disabling SSID broadcasting to make the network less visible to potential attackers.
Segmenting the wireless network into separate virtual LANs (VLANs) based on user
roles or departments adds an extra layer of security by restricting the lateral movement
of threats within the network.
Regular monitoring and intrusion detection play a crucial role in identifying and
mitigating security threats in real-time. Intrusion detection systems (IDS) or intrusion
prevention systems (IPS) can analyze network traffic patterns and detect suspicious
activities or unauthorized access attempts. Additionally, deploying wireless intrusion
detection/prevention systems (WIDS/WIPS) helps monitor the airwaves for rogue access
points, unauthorized clients, or other malicious activities that could compromise network
security.
Employee awareness and training are also essential components of a comprehensive
wireless network security strategy. Educating users about the risks of connecting to
unsecured wireless networks, the importance of using strong passwords, and the dangers
of phishing attacks can significantly reduce the likelihood of security incidents resulting
from human error or negligence.
Finally, regular security audits, vulnerability assessments, and penetration testing
should be conducted to identify weaknesses in the wireless network infrastructure
and address them promptly. Updating firmware, patches, and security configurations
for wireless devices and access points is crucial for staying ahead of emerging threats
and vulnerabilities.
INTRODUCTION
A few advantages that wireless communications provide to users and organizations are
enhanced productivity, reduced installation costs, and portability and flexibility. With
a wide range of applications and needs in mind, wireless technologies offer a diverse
range of capabilities. The threats to wireless network security are categorized into nine
groups in this chapter: malicious hackers, industrial espionage, malicious code, threats
to personal privacy, employee sabotage, fraud and theft by authorized or unauthorized
users of the system, loss of physical and infrastructure support, and errors and omissions.
Potential risks to wireless networks are all represented by the aforementioned. But the
immediate threats to wireless communications are industrial and foreign espionage, fraud
and theft, hostile hackers, and malicious code. Given how portable wireless devices are,
theft is more likely to happen. Fraud and theft can be committed by both authorized
and unauthorized users of the system, though the former are more likely to do so. It is
simpler for users to communicate when they are aware of the resources and security
vulnerabilities in the system.

WIRELESS NETWORK SECURITY 259

7.1. SYNOPSIS OF
WIRELESS TECHNOLOGY
→Learning Objectives

• Understand the fundamentals of wireless communication technologies

• Recognize various types of wireless devices such as smartphones, tablets,
laptops, and IoT devices
• Understand the risks associated with unsecured wireless networks and
unauthorized access
• Explore the latest trends and advancements in wireless communication
technologies.
• Gain an understanding of FIPS and its significance in information security
standards

In the simplest terms, wireless technologies allow for the communication of one or more
devices without the need for network or peripheral cabling or physical connections.
Whereas wired technologies use cables, wireless technologies use radio frequency
transmissions to transfer data. Wireless technologies cover a broad spectrum, from
sophisticated systems like wireless local area networks (WLAN) and cell phones to
basic gadgets like wireless microphones, headphones, and other non-processing, non-
storing devices. They also include wireless hi-fi stereo headsets, some cordless computer
keyboards and mice, and remote controls that use infrared (IR) technology. All of these
devices require a direct line of sight to break the connection.

7.1.1. Wireless Networks

A wireless network is one that is set up so that computers and other network devices
can communicate with one another using radio signal frequency. It is also known as
a WLAN or WiFi network at times. These days, this network is becoming more and
more popular because it requires no cabling and is simple to set up. Without the use
of cables, computers can be connected to any location in your house.

260 COMPUTER NETWORKS

 WIRELESS NETWORK SECURITY 261
Here’s a brief explanation of how it operates: Suppose you
have a wireless router installed and two PCs, each with a wireless
adapter. The binary data will be converted to radio frequency and
sent via wireless router when the computer sends out the data.
After that, the signal will be decoded back to binary data by the
receiving computer. Using a wireless network, you can access the
internet via either a DSL modem or a broadband cable. If you’ve
heard of a wireless hotspot, it’s a place where you and other
people can connect to a wireless network using wireless devices.
The two main components are wireless router or access point
and wireless clients. Simply purchase a wireless router and connect
it to your cable or DSL modem if you haven’t already set up a
wired network. After that, you create a basic wireless network
by installing wireless cards into each computer and configuring
wireless clients.

7.1.1.1. Wireless LANs

A wireless local area network, also known as a wireless LAN, is
made up of two or more computers connected to the same radio
frequency. This is in contrast to a wired LAN, where every computer
on the network is physically connected to the server’s network Keyword
switch or hub via an Ethernet cable. Every network starts with a Multiple-Input
primary computer, also called a server, and a number of connected Multiple-Output
(MIMO) is a
devices, also referred to as clients. Two Ethernet network interface wireless technology
cards (NICs) and network-supporting software are usually installed that uses multiple
on the server. Since Windows 98 Special Edition (SE), Microsoft transmitters and
receivers to transfer
Windows operating systems have included built-in networking more data at the
capabilities; however, third-party networking software is also same time.
accessible. A desktop computer could be the server and a laptop
the client in a basic home wireless LAN. Assume for the moment
that the desktop computer is connected to a high-speed digital
subscriber line (DSL) for Internet access. The desktop computer
must be physically linked to a wireless DSL modem in order to
share the Internet connection via a wireless local area network
(WLAN). A network switch or router may also be integrated into
the wireless DSL modem. These two devices maintain data flow to
the appropriate network machine. Should they not be integrated
into the modem, they will need to be bought independently.
Physical Ethernet cables are used to connect the desktop
server to the DSL modem, switch, and router. But the clients don’t
need any cabling. Instead, a wireless NIC needs to be installed
on every computer. This could be an internal wireless NIC, a USB
device, or a PCMCIA NIC. A tiny antenna is a feature of many
262 COMPUTER NETWORKS

wireless NICs, but not all of them. Radio designed specifically for that purpose. They
waves can be used by machines to send are frequently made quickly and for a single,
and receive data once the wireless LAN has fleeting purpose. These networks typically
been configured on the server and client. consist of a collection of workstations or
Because the client can stay mobile anywhere other wireless devices that exchange data
within the network’s broadcasting range, a directly with one another. Consider these
wireless LAN is incredibly convenient. With ties as impromptu networks that are open
a laptop, one can work from any room in to anyone in a particular location.
the house—mostly even the backyard—and
still share the server’s network connection. An ad hoc network is one in which no
access points allow participants to exchange
Mobile employees have instant access to data. Infrastructure networks route data
the office via a wireless LAN. Additionally, via a central information hub, which may
it saves money by eliminating the need be a computer’s hardware or software. For
to run Ethernet cables throughout an instance, office networks typically use a
entire building, enabling simple desktop server that business workstations connect
connectivity between clients. One of the to in order to obtain their information. On
key benefits of a wireless local area network the other hand, ad hoc networks bypass a
(LAN) is its ease of installation, as there central information hub.
are no wires needed to reach the clients. In
addition to being less expensive, wireless These networks are usually created
LANs offer greater flexibility than wired between participants and are closed
LANs. A wired network has fewer security networks, meaning they do not connect
measures to prevent intrusion, and wireless to the Internet. However, if one of the
LANs can be slower and less dependable participants is connected to a public or
when interference is present. These are the private network, other network members
two main drawbacks of wireless LANs. On may share this connection. This will make
the other hand, as technology advances, it possible for additional users on the
dependency rates also do. Certain wireless impromptu ad hoc network to establish an
local area networks (LANs) can rival wired Internet connection. Due to their ability to
networks. Wireless technology standards enable wireless gaming between players,
are denoted by letter designations. 54 ad hoc networks are frequently found
megabits per second are possiblwith the on portable gaming systems such as the
wireless “g” standard (mbps). The most Nintendo DS and Sony PSP. Customers can
recent iterations of multiple input multiple access new game demos through the store’s
output (MIMO) technology enable rates to own ad hoc network by creating networks
approach 100 mbps. within certain retail establishments. Peer-to-
peer networking for the wireless era is what
an ad hoc network is. Network environments
7.1.1.2. Ad Hoc Networks for early Windows computers were created
using peer-to-peer or workgroup style
Wireless networks have been classified into
networks. In a smaller office setting, this
two categories since their inception: ad
allowed the early computers to connect to
hoc networks and infrastructure networks,
each other and share information without
which include some LANs. The Latin term
the need for domains and the associated
“ad hoc” means “for this purpose.” Ad
extra overhead and management.
hoc networks, then, are networks that are
 WIRELESS NETWORK SECURITY 263
7.1.2. Wireless Devices Perhaps a Personal Area Network (PAN)
contains another kind of wireless device. A
Any type of communication equipment that
Personal Area Network (PAN) is established
can transmit data to another device without
through the use of Bluetooth® technology,
the need for a physical wire is referred to
which allows connections between personal
as a wireless device. Wireless headphones
digital devices over extremely short
with a receiver connect to a transmitter that
distances—a few feet, on average—though
is connected to the sound source, such as a
the standard is 30 feet (9.14 m). Networks
television, using either radio frequency (RF)
of the Bluetooth® variety are incredibly
or infrared technology. Generally speaking,
versatile and practical. It saves the trouble
though, a wireless device is a networking
of setting up shared resources over a LAN
device that can transmit data to other
by allowing print jobs to be sent from a
wireless network equipment even when it
laptop to a nearby printer. Moreover, it is
is not physically connected.
employed to link Apple products, PDAs, and
There are many different kinds of wireless cell phones with Bluetooth® capability to
technologies in today’s world, where people one another as well as to other Bluetooth®-
value having constant access to the Internet enabled devices like computers, headsets,
and to one another. Local area networks and standalone speakers. A Bluetooth®
(LANs) are broadcast by wireless routers network can operate “within” a LAN
with integrated modems, hubs, and switches without interfering since it operates on a
in homes and offices, enabling nearby different frequency spectrum than LANs.
computers to connect. A LAN typically
spans 300 feet (91.44 m) or more, though the 7.1.1.3. Personal Digital Assistants
actual broadcasting distance varies greatly
depending on numerous factors. A printer, A portable electronic gadget used for
other office supplies, or a fast Internet communication and personal organization
connection are just a few examples of the is called a personal digital assistant (PDA).
network resources that any computer on PDAs typically have memory cards for
the network can share. A wireless network removable storage, touch screens that allow
card or adapter needs to be installed on for finger or stylus input, organizational
a computer in order for it to connect to software like an address book and calendar,
a wireless LAN (WLAN). A network card and ways to connect to a personal computer
is an internal wireless device designed to via a cable or a wireless protocol like
operate using the same protocol or language Bluetooth®. The majority allow you to add
as wireless routers. Nevertheless, in the apps, like games or business software.
interim, compatibility problems arise as
these protocols sporadically develop into PDAs come in a wide variety, occupying
new standards. In cases where an internal a space between laptop computers and cell
wireless device is unable to support the phones in terms of both functionality and
protocol used by the router, an external size. Typically, they are compact enough
wireless adapter can be inserted into an to fit easily in a pocket or belt loop. A
external port. Although ExpressCard® smartphone is a PDA that has cell phone
formats are also available for wireless functionality; these are by far the most
network adapters, laptop users have an common kind of PDAs available.
additional option in addition to the more One of the key benefits of PDAs is their
popular USB dongle. capacity to replace numerous heavy paper
264 COMPUTER NETWORKS

organizational tools with one handy device. A software suite used

for personal organization is referred to as a “personal information
manager” (PIM). While some PIMs run on PCs or other platforms,
this suite is typically included with a personal digital assistant. A
calendar, address book, notepad, calculator, to-do list, and other
Remember small apps resembling those in a paper-based personal organizer are
Restrict access commonly included in this software. Software for spreadsheets and
by MAC address. word processing may also be included in more advanced bundles.
Many will Earlier on in the history of the device class, PDAs themselves were
tell you that
MAC address also referred to as “Personal Information Managers.””
restriction
doesn’t provide A number of operating systems (OS), called mobile operating
real protection systems, have been created specifically for PDAs. Many mobile
but, like hiding operating systems are based on familiar computer operating systems
your wireless — such as Microsoft Windows®, Mac OS®, or Linux® — that
network’s SSID,
restricting the
have been heavily modified to run using a mobile device’s limited
MAC addresses hardware and interface elements. Others are built from scratch with
allowed to the electronic device in mind. While the PC market has historically
connect to the been dominated by Windows®, no single OS holds a clear majority
network helps
in the mobile market. When personal digital assistants rose to
ensure you are
not one of the popularity in the late 1990s and early 2000s, many popular devices
“low hanging by Handspring™ and Palm®, Inc. ran Palm OS®, which featured
fruits” that a simple interface and primitive handwriting recognition software
people prefer to called Graffiti™. With the introduction of the Palm Pre™, the
attack.
operating system was succeeded by Palm webOS™. Microsoft’s
mobile OS is called Windows Mobile®; Apple’s is called iPhone®
OS. Symbian™ OS and the Google-initiated Android™ are other
mobile operating systems.
A minor backlash emerged in popular culture as a result of
the personal digital assistant’s quick ascent to fame and early
connotations as a status symbol for the business class. The cheeky
Hipster PDA was designed by self-help author Merlin Mann. It is
a set of index cards with a ballpoint pen and binder clips holding
them together. A user base has emerged to support the “device”
as a result of the concept becoming popular with some people.

7.1.1.4. Smart Phones

In addition to the basic function of placing calls, smartphones
offer additional features. A smartphone is typically thought of as
a mobile phone rather than a landline, even though the term can
be appropriately used to refer to any kind of phone. As mobile
devices have advanced in sophistication over time, the idea of the
“smartphone” has grown. In 1992, IBM created the first mobile
phone that could be categorized as a smartphone. Renamed Simon,
 WIRELESS NETWORK SECURITY 265
the gadget debuted at COMDEX that year theft of devices, denial of service attacks,
and went on sale to the general public the hostile hackers, malicious code, theft of
following year. The functionality included service, and foreign and industrial espionage.
several useful features previously exclusive Because wireless devices are so portable,
to desktop and laptop computers. Simon theft is more likely to happen. The system’s
added an email feature, a basic notepad, a users, both authorized and unauthorized,
calculator, and fax and email capabilities. have the potential to commit fraud and
Not too long after, the user could also play theft, but authorized users are more likely
a few games on the device and view the to do so. Users find it easier to commit
current times in major cities across the fraud and theft because they may be aware
globe with the help of a handy world clock. of the resources available to them and the
There were several kinds of cutting-edge security holes in the system. Hackers who
mobile phone devices available on the enter a system without authorization with
market for the balance of the 1990s. On malicious intent, also known as crackers,
many smartphones, touch screens started typically do so for personal gain or malicious
to take the place of keypads. New model intent. Typically, malicious hackers are
releases brought with them increasingly people who are not affiliated with any
complex email and gaming features. The agency or organization, though they can
quality of the speakerphone feature on many also be users there. Through wireless device
smartphone models started to compete with eavesdropping, these hackers may be able
that of handset speakerphones. A feature-rich to access the wireless network access point.
communication tool, the smartphone serves
many purposes today. New developments If successful, attacks stemming from
in smartphone technology include the these threats jeopardize an agency’s
addition of Internet access. Nowadays, systems and, more crucially, its data. The
people can browse the web just as easily main goals of all government security
as they would on a desktop or laptop. practices and policies are to ensure
The clarity and integrity of the phone’s availability, confidentiality, integrity,
basic audio signal have been improved and authenticity. Information must be
by numerous smartphone manufacturers shielded from unauthorized, unexpected,
concurrently. This makes it possible to use or inadvertent alteration, according to NIST
the smartphone to make a basic phone call Special Publication (SP) 800-26, Security
and still expect clear, sharp sound quality, Self-Assessment Guide for Information
even with all the extra features added. Technology Systems. The following are
among the requirements for security:
• Authenticity: A third party must be
7.1.3. Wireless Security Threats
able to verify that the content of
and Risk Mitigation a message has not been changed
Security risks are generally divided into nine in transit.
categories by An Introduction to Computer • Nonrepudiation: The origin or the
Security, which range from mistakes and receipt of a specific message must
omissions to threats to individual privacy. be verifiable by a third party.
These all stand for possible dangers to
• Accountability: The actions of an
wireless networks. But the more pressing
entity must be traceable uniquely
issues with wireless communications are
to that entity.
266 COMPUTER NETWORKS

Network availability is “the property of computer or voice (IP telephony)

being accessible and usable upon demand network through wireless
by an authorized entity.” connections, potentially bypassing
any firewall protections.
The information technology resource
(system or data) must be available on a • Sensitive information that is not
timely basis to meet mission requirements encrypted (or that is encrypted with
or to avoid substantial losses. Availability poor cryptographic techniques) and
also includes ensuring that resources are that is transmitted between two
used only for intended purposes. wireless devices may be intercepted
and disclosed.
The risks associated with operating a • Denial of service (DoS) attacks may
wireless network are the same as those be directed at wireless connections
associated with operating a wired network or devices.
(or any network, for that matter) plus the
• Malicious entities may steal the
additional risks brought about by flaws in
identity of legitimate users and
wireless protocols. Agencies must implement
masquerade as them on internal
security procedures and policies that help
or external corporate networks.
reduce these risks to a manageable level
in order to mitigate them. For example, • Sensitive data may be corrupted
before implementing wireless networks, during improper synchronization.
they must conduct security assessments to • Malicious entities may be able to
identify the specific risks and weaknesses violate the privacy of legitimate
that these networks will bring into their users and be able to track their
environments. They should take into account physical movements.
current security policies, known threats • Malicious entities may deploy
and vulnerabilities, laws and regulations, unauthorized equipment (e.g.,
safety, dependability, system performance, client devices and access points)
the life-cycle costs of security measures, and to surreptitiously gain access to
technological requirements when conducting sensitive information.
the assessment. Following the completion of
the risk assessment, the agency can start • Handheld devices are easily stolen
organizing and putting into practice the and can reveal sensitive information.
safeguards it will use to protect its systems • Data may be extracted without
and reduce security risks to a manageable detection from improperly
level. Because malicious threats and computer configured devices.
technologies are always evolving, the agency • Viruses or other malicious code may
should regularly review the policies and corrupt data on a wireless device
procedures it has put in place. (For further and be subsequently introduced to
in-depth details regarding risk mitigation a wired network connection.
and.)
• Malicious entities may, through
• All the vulnerabilities that exist in wireless connections, connect to
a conventional wired network apply other agencies for the purposes of
to wireless technologies. launching attacks and concealing
their activity.
• Malicious entities may gain
unauthorized access to an agency’s • Interlopers, from inside or out,
 WIRELESS NETWORK SECURITY 267
may be able to gain connectivity to network management
controls and thereby disable or disrupt operations.
Malicious entities may use a third-party, untrusted wireless
network services to gain access to an agency’s network resources.
Internal attacks may be possible via ad hoc transmissions.
Agency representatives must be aware of liability concerns
for the loss of confidential data or for any attacks coming from a
compromised network, just like with wired networks.

7.1.4. Emerging Wireless Technologies

Today, with the global mobile traffic expected to increase nearly
11 times between 2013 and 2018, there is a strong need for more
efficient use of bandwidth and faster mobile speeds. The market
is being dominated by smart devices with high processing power
and network connectivity; there are currently 7 billion wireless
devices in use. With the advent of the Internet of Things (IoT),
Keyword
where smart sensing and machine-to-machine communication are Internet of
Things (IoT)
expected to skyrocket in the next ten years (with applications to is a name for the
smart environments, smart health, etc.), this number is expected to aggregate collection
rise by an order of magnitude. ( ). All of this indicates that there of network-enabled
devices, excluding
are fundamentally new problems that need to be solved using traditional
knowledge from information theory, computer science, mathematics, computers like
and economics. laptops and servers.

A new wireless standard called “5G” will be developed over the

next five years with the goal of increasing system capacity by orders
of magnitude in order to address these issues. Current technologies
will not be sufficient to enable this, and fundamentally new ideas
must be developed to deal with the anticipated proliferation of
IoT devices. In this workshop, basic new concepts in wireless
networks and their mathematical applications will be examined.
While there are numerous conferences and workshops focused on
wireless systems and implementations, none that we are aware
of will bring together traditionally unrelated fields like computer
science, applied mathematics, information theory, and wireless
networks. Researchers working on a number of fundamental topics
that could have a significant influence on wireless networks in
the future will be brought together for this workshop. Information
theoretic and entropy inequalities, coding theory, probabilistic
analysis—which includes the analysis of (randomized) algorithms—
convex optimization, stochastic geometry, random matrices, and
other mathematical tools will all be used.
268 COMPUTER NETWORKS

7.1.5. Federal Information communications and data used in electronic

commerce, critical infrastructures, and
Processing Standards
other application areas. These systems and
FIPS 140-2 defines a framework and products use cryptographic modules to offer
methodology for NIST’s current and future cryptographic functions like identification,
cryptographic standards. The standard authentication, confidentiality, integrity, and
provides users with the following: nonrepudiation. For security assurance, the
• A specification of security features cryptographic module must be adequately
that are required at each of four tested and validated against accepted
security levels standards. The public and federal agencies
alike gain from the use of products that
• Flexibility in choosing security have been verified and tested. Insufficient
requirements testing can lead to insecure products
• A guide to ensuring that the due to flaws like shoddy design, subpar
cryptographic modules incorporate algorithms, or improper cryptographic
necessary security features module implementation.
• The assurance that the modules are
The Cryptographic Module Validation
compliant with cryptography-based
Program (CMVP), developed by NIST in 1995,
standards.
verifies cryptographic modules in accordance
According to the Secretary of Commerce, with FIPS 140-2, Security Requirements for
FIPS 140-2 is now required and enforceable Cryptographic Modules, and other standards
for U.S. government organizations. When a based on FIPS cryptography. NIST and the
federal agency concludes that cryptography Canadian government’s Communications
is required to protect sensitive data, the Security Establishment (CSE) collaborated
standard is particularly applicable. The to create the CMVP. The federal agencies of
standard is applied in the design and both nations accept products that have been
implementation of cryptographic modules verified to comply with FIPS 140-2 for the
that are operated, or have been operated, protection of sensitive data. Cryptography
by federal departments and agencies. FIPS module vendors test their products in
140-2 is relevant whether the module is independent, approved testing facilities.
used as a standalone device or integrated As the program’s validation authorities,
into a product or application. Both Bluetooth NIST’s Computer Security Division and CSE
and 802.11 security as it exists now do not work together to verify the test findings.
adhere to the FIPS 140-2 standard. FIPS 140-2 compliance testing is currently
carried out by six laboratories that have
Cryptography is used by federal agencies,
been accredited by the National Voluntary
businesses, and the general public to secure
Laboratory Accreditation Program (NVLAP).

Did you get it?

1. Explain the importance of device compatibility and interoperability in wireless networks.
2. Discuss the implications of adopting emerging wireless technologies on network performance,
security, and scalability.
 WIRELESS NETWORK SECURITY 269

7.2. DEVELOPMENT OF
WIRELESS LANS
→Learning Objectives
• Explore the role of WLAN in providing flexible and mobile connectivity within
organizations.
• Discuss the scalability and flexibility offered by WLAN for accommodating
changing business needs.
• Learn about the security mechanisms and protocols available for securing 802.11
Wireless LANs.
• Identify the security requirements and compliance standards applicable to
WLAN deployments.
• Explore the latest advancements in WLAN security standards and technologies.

A wireless LAN, or wireless local area network, consists of two or more computers
linked through the same radio frequency. In contrast, a wired LAN requires physical
connections via Ethernet cables to a server’s network switch or hub. Each network
begins with a main computer, known as a server, and multiple connected devices, called
clients. Two Ethernet network interface cards (NICs) and network-supporting software
are usually installed on the server. Since Windows 98 Special Edition (SE), Microsoft
Windows operating systems have included built-in networking capabilities; however,
third-party networking software is also accessible. A desktop computer could be the
server and a laptop the client in a basic home wireless LAN.
Imagine if the desktop computer is hooked up to a high-speed DSL for Internet
access. To spread the Internet connection through a wireless LAN, the desktop computer
needs to be physically connected to a wireless DSL modem. The wireless DSL modem
might have a network switch or router built in to keep the data flowing to the right
network device. If these aren’t part of the modem, you’ll have to get them separately.
Physical Ethernet cables connect the desktop server to the DSL modem, switch, and
router. Clients require wireless NICs like internal, USB, or PCMCIA. Many wireless NICs
have a small antenna.
With a laptop, one can work from any room in the house—mostly even the backyard—
and still share the server’s network connection. Mobile staff members can instantly
connect to the office via a wireless LAN. Additionally, it provides simple, easy desktop
connectivity between clients without incurring the high cost of running Ethernet cable
throughout a building. One of the primary benefits of a wireless LAN is its ease of
270 COMPUTER NETWORKS

installation, as there are no wires leading to the clients. In addition

to being less costly than wired LANs, wireless LANs offer greater
flexibility. Wireless local area networks (LANs) have two drawbacks:
they can be slower and less dependable than wired networks when
interference is present, and they require additional steps to secure
Remember against intrusion. However, as technology advances, dependency
rates also tend to improve. Certain wireless LAN configurations
Radio waves can
be used by the
are competitive with wired networks.
machines to send
and receive data
Wireless technology standards use letter designations. The “g”
once the wireless standard allows speeds of up to 54 mbps. Recent MIMO technology
LAN has been versions can achieve rates close to 100 mbps. This corresponds to
configured on the speed of a typical hard-wired network and DSL connection.
the server and
The slower technologies of the earlier wireless LAN standards “a”
client. Because
the client can and “b”.
stay mobile
anywhere within
the network’s 7.2.1. Purpose of Wireless LAN
broadcasting
range, a wireless The Federal Communications Commission (FCC) initially opened up
LAN is incredibly the RF spectrum to industry in the mid-1980s, which is when WLAN
convenient. technology and the WLAN sector began. Growth was comparatively
slow in the 1980s and early 1990s. But WLAN technology is growing
at a very rapid rate right now. The more bandwidth made available
by the IEEE 802.11 standard is the main driver of this expansion.
Table 7.1 gives an overview of some of the most important features
of WLAN and 802.11 technology.
Table 7.1. Key Characteristics of 802.11 Wireless LANs

Characteristic Description
Physical Layer Direct Sequence Spread Spectrum (DSSS), Frequency
Hopping Spread Spectrum (FHSS), Orthogonal Frequency
Division Multiplexing (OFDM), infrared (IR).
Frequency Band 2.4 GHz (ISM band) and 5 GHz.
Data Rates 1 Mbps, 2 Mbps, 5.5. Mbps (11b), 11 Mbps (11b), 54
Mbps (11a )
Data and Network RC4-based stream encryption algorithm for confidentiality,
Security authentication, and integrity. Limited key management.
(AES is being considered for 802.11i.)
Operating Range Up to 150 feet indoors and 1500 feet outdoors.9
Positive Aspects Ethernet speeds without wires; many different products
from many different companies. Wireless client cards and
access point costs are decreasing.
Negative Aspects Poor security in native mode; throughput decrease with
distance and load.
 WIRELESS NETWORK SECURITY 271
7.2.1.1. Frequency and Data Rates hoc networks. In an infrastructure BSS, all
communications—including those between
In order to provide wireless networking mobile nodes within the same service set—
technology similar to the long-established take place over APs. The coverage distance
wired Ethernet, IEEE developed the 802.11 from the AP defines the boundaries of an
standards. The 802.11 WLAN family’s most infrastructure BSS. A single AP’s coverage
extensively used standard is IEEE 802.11a. area is referred to as a cell. The AP needs
It uses OFDM technology and operates in to be able to reach every mobile station.
the licensed 5 GHz band. The widely used
802.11b standard uses direct sequence
spread-spectrum technology to function in
the unlicensed 2.4. GHz–2.5. GHz Industrial,
Scientific, and Medical (ISM) frequency
band. Since the ISM band is accessible
everywhere, it has grown in popularity for
wireless communications. Up to 11 Mbits
per second of transmission are possible with
the 802.11b WLAN technology. This makes
it marginally faster than regular Ethernet
and much faster than the original IEEE
802.11 standard, which transmits data at
up to 2 Mbps.

7.2.1.2. 802.11 Architecture

Figure 7.1. Independent and Infrastructure
The Basic Service Set (BSS) and the Extended Basic Service Sets (Source: ScienceDirect,
Service Set (ESS) are the two types of https://ars.els-cdn.com/content/image/3-s2.0-
B9781856177474000056-gr3.jpg).
services that are defined by the IEEE 802.11
standard. The fundamental component of a 802.11 enables the creation of wireless
wireless LAN is the BSS. A BSS is made up networks with arbitrarily large coverage
of mobile or fixed wireless stations, as well by connecting BSSs to an ESS. BSSs are
as maybe a central base station (e.g., an connected via a backbone network to form
AP). A station can communicate with other an ESS. An ESS’s Service Set Identifier
BSS members when it is a part of the BSS. (SSID), which acts as a network name for its
A BSS without an AP is an isolated users, is assigned to each AP in the system.
network that is unable to communicate The way APs in an ESS function allows
with other BSSs. We refer to these BSSs the outside world to communicate with
as Independent BSSs (IBSS). IBSSs typically a station using its MAC address without
consist of a limited number of stations that having to worry about where it is in the
are set up for a particular function and for ESS. Three BSSs that correspond to three
a brief amount of time (e.g., establishing APs are shown in Figure 7.2. Between BSS
a transient network to facilitate a solitary 2 and BSS 3, as well as between BSS 1 and
conference room meeting). Another name for BSS 2, there is an equal amount of overlap.
IBSSs is ad hoc networks. The usage of an In order to ensure that stations can move
AP sets infrastructure BSSs apart from ad seamlessly from one BSS to another, this
overlap is required. In the figure, frames
272 COMPUTER NETWORKS

are delivered to a station by the router using its MAC address as the destination; only
the associated AP can deliver the frame to that station.

Figure 7.2. Extended service set (Source: Geeksforgeeks, https://media.geeksforgeeks.org/wp-

content/uploads/20201105165935/fffd1.png).

Typically, the main driver behind the deployment of an 802.11 network is mobility
support. At the link layer, IEEE 802.11 permits mobility between BSSs. It is ignorant
of all activity occurring above the link layer, though. Stations will search for and try to
associate with an AP with the strongest signal and the least amount of network traffic
when they move between BSSs. In this manner, a mobile station can move between
APs in the network without experiencing a loss of connectivity. This phenomenon is
commonly known as roaming.

7.2.1.3. Wireless LAN Components

A wireless station and an access point are the two pieces of hardware that make
up a WLAN. A laptop or notebook PC with a wireless network interface card (NIC)
is commonly referred to as a station, or client. Additionally, a desktop computer or a
portable electronic device (e.g., PDA, or specialized gadget like a barcode scanner) or
hardware found in a kiosk on a production line or other place that is open to the public.
With the exception of using wireless NICs to connect to network access points, wireless
laptops and notebooks, also known as “wireless enabled,” are identical to laptops and
notebooks. Usually, the wireless NIC is installed in the client.
A USB port or a PCMCIA slot is a type of connector used for memory cards in
computers. In order to connect to the WLAN, the NICs use radio signals. Usually
made up of a radio, a wired network interface (like 802.3), and bridging software, the
access point (AP) serves as a bridge between wireless and wired networks. The AP
aggregates several wireless stations onto the wired network, acting as a base station
for the wireless network.
 WIRELESS NETWORK SECURITY 273
7.2.1.4. Range not at home or at work. One needs the
required hardware in order to use a wireless
The sharing of data between wireless connection to access the Internet. Almost
devices or computers is known as long-range all laptops manufactured after the year
wireless. Wireless technology, commonly 2000 have a wireless transmitter installed.
referred to as WiFi, enables users to connect Users typically have the option of buying
from up to 40 miles away (about 64.37 km) a wireless adapter for desktop and laptop
and transmits data at high speeds. Computer computers that might not be wirelessly
data is transformed into radio waves and compatible. Wireless adapters are made to
sent through the air in wireless technology. fit into PC card slots and Universal Serial
These waves are detected by a wireless Bus (USB) ports.
router, which converts them back into
data. The technology behind this process
is comparable to that of walkie-talkies, cell 7.2.2. Benefits
phones, and other two-way communication WLANs offer four primary benefits:
devices. Long-range wireless has a high
speed that allows for the transfer of more • User Mobility: Users can access
data in less time. Even though wireless files, network resources, and
technology is quick, there are situations the Internet without having to
when even faster performance is required. physically connect to the network
Typically, power amplifiers or specialized with wires. Users can be mobile yet
antennas can increase the connection speed retain high-speed, real-time access
and range. to the enterprise LAN.
However, adding boosters can have a • Rapid Installation: The time required
drawback in that they not only increase for installation is reduced because
speed but also increase the connection’s network connections can be made
susceptibility to threats and other technical without moving or adding wires,
issues. The obstruction of line of sight (LOS) or pulling them through walls or
by long-range wireless is a minor additional ceilings, or making modifications to
concern. This means that users in cities the infrastructure cable plant. For
have to take into account obstacles like example, WLANs are often cited as
radio stations, buildings, and microwave making LAN installations possible
signals, while residents in more rural areas in buildings that are subject to
are more likely to be concerned about historic preservation rules.
trees and signal loss from being too far • Flexibility: Enterprises can also
outside of the connection range. Millions of enjoy the flexibility of installing and
people use wireless technology worldwide. taking down WLANs in locations as
Wireless technology is used by companies necessary. Users can quickly install
and educational institutions for coverage a small WLAN for temporary needs
establishment, remote support, and research such as a conference, trade show,
completion. Long-range wireless is used or standards meeting.
in homes for local WiFi networks, remote • Scalability: WLAN network
access, and hard-to-reach rural locations. topologies can easily be configured
There are public WiFi hotspots where to meet specific application and
people can use a wireless transmitter installation needs and to scale
to connect to the Internet when they’re
274 COMPUTER NETWORKS

from small peer-to-peer networks network traffic, nor does it offer any other
to very large enterprise networks protection other than access control to the
that enable roaming over a broad network behind the access point, as its name
area. suggests. As a result, any wireless network
The WLAN market has been growing that is only secured by this method is
steadily over the past few years, and WLANs extremely weak and susceptible to network
are still becoming more and more popular, attacks. Furthermore, it turns out that it
thanks to these core advantages. WLANs is relatively simple to temporarily change
are starting to show promise as a substitute the MAC address on almost any wireless
for conventional wired solutions. To carry network interface (at least in Windows and
out their regular business operations, UNIX/Linux OS), rendering this security
hospitals, colleges, airports, hotels, and mechanism useless as an access control
retail stores, for instance, already use measure. MAC spoofing is the process of
wireless technologies. altering the MAC address to pretend to be
another person or device. To perform MAC
spoofing in UNIX/Linux OS, we simply need
7.2.3. Security of 802.11 to run one command.
Wireless LANs ifconfig eth0 hw ether 00:01:02:03:04:05
It is very difficult to control which
computers or devices are receiving the Where:
wireless network signal because of the RF eth0 is the network interface we want
signal nature of the network. In order to to modify.
prevent eavesdropping and other network
threats, software link-level protection— 00:01:02:03:04:05 is the MAC address
which employs cryptography—is therefore we want to specify with numbers in hex
necessary for wireless. Exclusively WEP is number format.
available for wireless network security in
the original 802.11 standard. In Windows environment, we need the
help of tools designed specially for this
intention. There are at least two tools
7.2.3.1. Access Control List available for MAC spoofing:
In a wireless network, the most basic a. SMAC3, from KLC Consulting, and
security feature is the access control b. MAC Makeup4 (freeware), from
list. This mechanism provides protection H&C Works
mostly by filtering out unknown users and
necessitates loading the MAC addresses Both feature an intuitive graphical user
of authorized clients into the access interface, work with Windows XP and 2000,
point. The Access Point will only accept don’t require drivers to be installed, and
communication from those MAC addresses work almost with any network interface
that have registered; any communication card—including wireless ones. Anyone can
from unregistered MAC addresses will be obtain a list of registered MAC addresses
rejected. on their own with the use of wireless
analyzer utilities, which enables anyone
This mechanism does not protect the to use this information to gain access to
confidentiality or integrity of wireless an unprotected wireless network. With the
 WIRELESS NETWORK SECURITY 275
ability to passively monitor or sniff wireless network traffic, wireless analyzers are
incredibly potent tools for learning vital details about wireless networks.

Did you know?

Norman Abramson, a professor at the University of Hawaii, developed the world’s first wireless computer
communication network, ALOHAnet. The system became operational in 1971 and included seven computers
deployed over four islands to communicate with the central computer on the Oahu island without using phone lines.

7.2.3.2. WEP (Wired Equivalent Privacy)

At first, the 802.11 standard only defined WEP (Wired Equivalent Privacy) as a link-
level security option. Its primary goal was to safeguard the integrity and confidentiality
of wireless network traffic. WEP gets its name from the fact that it was intended to
offer similar secrecy to a conventional wired network. To achieve the required level of
security, data is encrypted using WEP. To encrypt data packets, WEP employs the RC4
stream cipher with a 64- or 128-bit key. WEP can also be used as an access control
mechanism because, once enabled, it only allows communication between the Access
Point and nodes that have a shared secret key, rejecting all other nodes.

Figure 7.3. WEP Encryption/Decryption Process. (Black) Common steps, (Red) Encryption Process,
(Blue) Decryption Process (Source: Researchgate, https://www.researchgate.net/profile/Abdel-
Karim-Al-Tamimi/publication/303417816/figure/fig9/AS:627929046335490@1526721359103/WEP-
encryption-and-decryption-processes.png).

Every wireless client on a WEP-protected network needs to have the same secret
key as the access point in order to function. This secret key may take the form of a
passphrase or a string of characters produced by a wireless configuration software.
The most crucial requirement for every member of the wireless network is that they
all have the same WEP secret key. Although WEP uses a 64- or 128-bit key to encrypt
data, the actual effective key is smaller because a portion of the WEP key is transmitted
in clear text along with the data packet. This is especially true when using cards
276 COMPUTER NETWORKS

from multiple manufacturers. The shared of the first logical XOR operation. This is
secret key (the static portion of the key) possible because a specific WEP key (the
and the dynamic Initialization Vector (IV) same IV and shared secret key) will always
value are concatenated to form the WEP produce the same pseudo-random string,
key, which is the key used to encrypt the making it possible to regenerate the same
data packet. In order to give the WEP key pseudo-random string.
liveness, the transmitter wireless network
interface selects the Initialization Vector When the WEP was designed, it was
(IV), a dynamic 24-bit value that offers more thought to be sufficiently secure. However,
than 16 million possible keys at random. later research revealed an inherent
WEP keys must be live since every message weakness in the WEP mechanism. Since
needs to be encrypted using a unique key. then, a number of tools have been created
The duration of the mutual. to successfully crack the WEP shared secret
key, making it unsecure.
A data packet that reaches the link level
in a wireless network protected by WEP is
7.2.3.3. WEP Key Recovery
encrypted before being sent over the air.
The wireless network interface first selects A serious cryptographic flaw in the WEP
an IV value at random and concatenates it protocol makes it possible for an attacker
with the shared secret key to create the to obtain the shared secret key. This flaw
WEP Key (IV + secret key). This is how involves taking advantage of the fact that a
data packet encryption is carried out. The specific IV value generates weak WEP keys.
selection criteria for the wireless network The first bytes of the pseudo-random string
interface are not specified in the 802.11 may have some correlation with the WEP key
standard. As a result, the manufacturer when encryption is performed with a weak
determines how the wireless network WEP key. There is a 5-6 percent certainty
interface determines the IV value. When probability that each weak key will leak one
the WEP Key is prepared, the RC4 stream byte of the shared secret key. The attacker
cipher uses it to generate a pseudo-random needs to collect a significant number of
string that matches the length of the data raw WEP-protected data packets (roughly
packet. When the wireless network interface 5 million to 10 million packets) in order
performs a logical XOR operation between to increase the likelihood that the shared
the data packet and the pseudo-random secret key will be successfully cracked
string, encryption occurs. The link-level because the certainty level is low. This flaw
headers, IV value, and encrypted data packet was revealed in the paper “Weaknesses in
are combined and sent to the recipient to the Key Scheduling Algorithm in RC4” by
complete the WEP-protected data packet. Fluhrer, Mantin, and Shamir.
The procedures for encrypting messages
and decrypting a data packet protected by A number of WEP cracking tools have
WEP are the same. The recipient reads the already been available for some time.
IV value before repeating the encryption Wepcrack and Airsnort8 are a couple of the
process when decrypting a WEP-protected first tools that take advantage of the FMS
data packet. The data is decrypted by vulnerability. Other tools, such as brute-
executing a second logical XOR operation force attacks and optimizations for requiring
between the encrypted data packet and the fewer weak WEP-protected data packets,
pseudo-random string, canceling the effect have more capabilities. Among the tools are
 WIRELESS NETWORK SECURITY 277
Dwepcrack, Aircrack, and Airjack. Although encryption, thereby addressing the weak
most of these tools are designed to operate key vulnerability. Reducing the number of
on UNIX/Linux, some have already been WEP keys by avoiding some IVs exacerbates
adapted for Windows OS. By passively the IV collision weakness.
capturing a significant amount of wireless
network traffic, attackers can potentially
7.2.3.5. Dynamic WEP
obtain the shared secret key by utilizing any
of these WEP cracking tools. The wireless An attacker must intercept a significant
network is completely vulnerable as soon number of data packets in order to decipher
as the attacker manages to get their hands the shared secret key before they can gain
on the shared secret key. physical access to a WEP-protected network.
It will be much more difficult to crack the
7.2.3.4. IV Collision secret key if the wireless network is able to
alter the shared secret key each time before
An additional method of getting into a the attacker receives enough data packets.
wireless network without the shared Therefore, regularly changing the shared
secret key is to passively capture a lot of secret key is a way to stop a potential
data packets, but this time the attacker secret key recovery without aggravating
is searching for IV collisions. When two any other vulnerability. Because the current
or more data packets are encrypted using WEP mechanism lacks an automatic key
the same IV value—and thus, the same rotation system that could support this
WEP key—this is known as an IV collision. solution, using it could be very problematic.
An attacker can remove encryption by
performing a logical XOR with the two To accomplish this, automatic key
encrypted data packets when an IV collision delivery and recurring rekeying can be
is detected. The XOR of the two data facilitated by utilizing the 802.1x protocol.
packets is the outcome. If the attacker has For user authentication and authorization,
sufficient time and employs analytical and the 802.1x protocol is responsible. When
statistical techniques, they may be able to a user is approved and authenticated, the
retrieve the contents of the two packets. authentication server generates the WEP
When more packets encrypted with the key..
exact same WEP key are intercepted and
used for the recovery, the effort and time 7.2.3.6. WPA (Wi-Fi Protected Access)
needed to recover the contents of the data
packets decrease. It should be noted that The Wi-Fi Alliance has released WPA as
recovering a pseudo-random string for a a solution while a final security protocol
specific WEP key is simple if this attack is being standardized. The next 802.11
is successful. After that, an attacker could wireless network security protocol standard,
store all of the pseudo-random strings in 802.11i, is the foundation of this security
a record and use them for network access, protocol. The three primary parts of WPA
packet forging, and decryption. are TKIP, 802.1x, and MIC. Every element
was created and put into place with a
Aware of the WEP key vulnerability, specific 802.11 vulnerability in mind.
some manufacturers of wireless interface
cards provide updates for their cards Significant security enhancements were
that prevent the use of weak IVs for put into place, like the 802.1x protocol
278 COMPUTER NETWORKS

for network access control and the key secret key is only used as a seed for creating
hierarchy that essentially neutralizes and additional keys; it is not used to access the
safeguards the WPA main key from attacks. cipher directly. This strategy reduces the
By using key hierarchy, WPA avoids using likelihood that any attack will reveal the
the main key for encryption directly. Rather, shared secret key. The session key is the
the main key (Pairwise Master Key) is used first key to be generated. The per-packet
to generate other temporal keys, like session key is generated using this key as a seed.
keys and group keys, and then the session
key is used recursively to generate the
7.2.3.8. Security Features of 802.11 Wireless
encryption key for each packet. In order
LANs per the Standard
to prevent replay attacks, the IV is also
extended from 24 to 48 bits in length and The three basic security services defined
given a new function as a sequence counter. by IEEE for the WLAN environment are
Using a specially created cryptographically as follows:
protected hashing function rather than the
CRC32 linear function improves packet • Authentication: A primary goal of
integrity protection. WEP was to provide a security
service to verify the identity of
communicating client stations.
7.2.3.7. TKIP (Temporal Key Integrity This provides access control to the
Protocol) network by denying access to client
This protocol fixes the majority of critical stations that cannot authenticate
vulnerabilities and is a direct replacement properly. This service addresses
for WEP. Maintaining compatibility with the question, “Are only authorized
current 802.11 hardware allowed for persons allowed to gain access to
software upgrades, which was part of its my network?”
intended purpose. The fact that each packet • Confidentiality: Confidentiality,
is generated by a per-packet key mixing or privacy, was a second goal of
function rather than by concatenating the WEP. It was developed to provide
IV and the shared secret key is one of the “privacy achieved by a wired
most significant improvements over WEP. network.” The intent was to prevent
information compromise from casual
Since the majority of cryptographic eavesdropping (passive attack).
features are hard-coded into the hardware This service, in general, addresses
of the wireless network interface, software the question, “Are only authorized
upgrades are not possible. Owing to these persons allowed to view my data?”
constraints, the WPA protocol makes use of
• Integrity: Another goal of WEP
certain hardcoded functions from the WEP
was a security service developed
protocol in order to maintain performance
to ensure that messages are not
and compatibility. Certain functions, such
modified in transit between the
as the output function of the RC4 stream
wireless clients and the access
cipher, are hardcoded into the wireless
point in an active attack. This
network interface and cannot be altered.
service addresses the question, “Is
TKIP reuses the RC4 stream cipher to
the data coming into or exiting the
circumvent this issue, but it modifies how
network trustworthy—has it been
the shared secret key is used. The shared
tampered with?”
 WIRELESS NETWORK SECURITY 279
It is important to note that the standard did not address other
security services such as audit, authorization, and nonrepudiation.
The security services offered by 802.11 are described in greater
detail below.

7.2.3.9. Authentication
Open-system authentication and shared-key authentication are
the two ways that wireless users trying to connect to a wired Keyword
network can be “validated” according to the IEEE 802.11 standard. Cryptography
Cryptography underpins one method of authentication (shared-key is the process of
hiding or coding
authentication), but not the other. information so that
only the person
a message was
intended for can
read it.

Figure 7.4.Taxonomy of 802.11 Authentication Techniques (Source:

Researchgate, https://www.researchgate.net/profile/Karen-Scarfone/
publication/238668444/figure/fig7/AS:484492956704776@1492523530652/1-
Taxonomy-for-Pre-RSN-and-RSN-Security.png).

Since the access point accepts the mobile station without

confirming the station’s identity, the open-system authentication
technique is not really authentication. Furthermore to be mentioned
is the fact that the authentication process is one-way, validating
only the mobile station. The mobile station needs to have faith that
the AP it is speaking to is legitimate. Figure 7.4 shows a taxonomy
of the methods for 802.11.
A client using Open System authentication is considered
authenticated if it merely provides its MAC address in response to
an access point’s two-message exchange. The client simply replies
with the appropriate fields in the message exchange; they are not
actually validated during the exchange. It is evident that open-system
authentication is highly vulnerable and essentially invites unwanted
access in the absence of cryptographic validation. According to
280 COMPUTER NETWORKS

the 802.11 specification, only open-system straightforward unilateral challenge-

authentication is necessary. response schemes have been around for
a while.

7.2.3.10. Privacy
The 802.11 standard uses cryptographic
methods for the wireless interface to
support privacy (confidentiality). The RC4
symmetrickey stream cipher algorithm is also
used by the WEP cryptographic technique
for confidentiality in order to produce a
pseudo-random data sequence. To the data
to be transmitted, this “key stream” is
Figure 7.5. Shared-key Authentication Message simply added modulo 2 (exclusive-OR-ed).
Flow (Source: Medium.com, https://miro.medium. Data can be shielded from disclosure during
com/v2/resize:fit:640/format:webp/1*olfHTS- transmission over a wireless link by using
hoqEtiSDIXxxipA.png). the WEP technique. To secure traffic such
One cryptographic method of as Transmission Control Protocol/Internet
authentication is shared key authentication. Protocol (TCP/IP), Internet Packet Exchange
It is a straightforward “challenge-response” (IPX), and Hyper Text Transfer Protocol
system that asks clients if they are aware (HTTP), WEP is applied to all data above
of a shared secret. In this scheme, which the 802.11 WLAN layers.
is conceptually shown in Figure 7.5, the According to the 802.11 standard,
wireless client receives a random challenge WEP can only support shared keys with
generated by the access point. The challenge, a cryptographic key size of 40 bits.
also known as a “nonce” in security jargon, Numerous vendors do, however, provide
is encrypted by the client using a shared nonstandard WEP extensions that support
cryptographic key with the AP, and the key lengths ranging from 40 to 104 bits.
client then sends the result back to the A keysize of 128 bits is supported by at
AP. The client computes a result, which the least one vendor. For example, a 128-bit
AP decrypts. Access is granted only if the RC4 key can be created from a 104-bit
decrypted value matches the transmitted WEP key with a 24-bit Initialization Vector
random challenge. The RC4 stream cipher, (IV). Generally speaking, a cryptographic
created by MIT’s Ron Rivest, is the technique’s security is increased when the
algorithm used for both the cryptographic key size is increased, barring any other
computation and the creation of the 128- circumstances. Long keys may not, however,
bit challenge text. It should be mentioned always be able to boost security due to
that the just-described authentication poor implementations or designs. Studies
method lacks mutual authentication and have indicated that when considering robust
is a simple cryptographic technique. In designs and implementations, brute-force
other words, there is no guarantee that cryptanalysis (code breaking) is rendered
a client is interacting with a genuine AP impossible for key sizes larger than 80 bits.
and wireless network since the client does The keyspace of over 1026 possible keys for
not authenticate the AP. Not to mention, 80-bit keys is larger than what is currently
 WIRELESS NETWORK SECURITY 281
possible to compute. The majority of WLAN deployments in use
today rely on 40-bit keys. Furthermore, regardless of keysize, recent
attacks have demonstrated that the WEP approach for privacy is
regrettably susceptible to some attacks.

7.2.3.11. Integrity
A method for guaranteeing data integrity for messages sent between
wireless clients and access points is also described in the IEEE
802.11 specification. The purpose of this security service was to
reject any messages that had been altered by a hostile party
“in the middle.” This method employs a basic encrypted Cyclic
Redundancy Check (CRC) methodology. A CRC-32, or frame check
sequence, is calculated on each payload before transmission, as
shown in the above diagram. The cipher-text message is then
obtained by encrypting the integrity-sealed packet using the RC4
key stream. Decryption and CRC recalculation are done on the
receiving end after the message has been received. The CRC
calculated with the original message and the one computed at
the receiving end are compared. The packet would be rejected if Remember
the CRCs do not equal, indicating that the message was “received IEEE 802.11 is
in error,” which would be an indication of an integrity violation used in most
caused by an active message spoofer. Unfortunately, regardless of home and office
key size, the 802.11 integrity is susceptible to some attacks, just networks to
like the privacy service. In conclusion, the basic weakness in the allow laptops,
printers,
WEP integrity scheme lies in the fact that the straightforward CRC smartphones, and
is not a “cryptographically secure” mechanism. other devices
to communicate
Unfortunately, key management—the life cycle handling of with each other
cryptographic keys and related materials—is not covered by the and access the
IEEE 802.11 specification. Thus, those using WLANs are responsible Internet without
connecting wires.
for creating, distributing, storing, loading, escrowing, archiving,
auditing, and destroying the content. The 802.11 network’s
users are primarily responsible for managing their keys, which
is arguably the most important component of any cryptography
system. Consequently, the WLAN environment may be exposed
to numerous vulnerabilities.
Weak keys (all zeros, all ones, based on easily guessed
passwords, or other similar trivial patterns) and non-unique,
factory-default, and never-changing WEP keys are some examples
of these vulnerabilities. WEP-secured WLANs also do not scale well
since key management was not addressed in the original 802.11
specification, leaving the key distribution unresolved. In a large
WLAN environment, the task is daunting if an enterprise realizes
it needs to frequently change keys and make them random. As an
282 COMPUTER NETWORKS

illustration, a sizable campus might have which allows multiple users to potentially
up to 15,000 APs. It is a significant task share the same key for extended periods
to generate, distribute, load, and manage of time. The WEP protocol’s absence of
keys in an environment this size. any important management clauses is
partially to blame for this. The key could
be compromised along with all the other
7.2.3.12. Problems with the IEEE 802.11
computers that share it if a laptop or other
Standard Security
computer were lost or stolen. Furthermore,
The 802.11 WLAN standard’s standardized if every station uses the same key, an
security has this vulnerability. WLANs eavesdropper may have quick access to
based on 802.11 employ the WEP protocol, a lot of traffic for analytic attacks like the
as previously mentioned. In response, WEP ones shown in numbers 2 and 3 below.
protects traffic using the RC4 cryptographic
The IV in WEP is a 24-bit field that
algorithm and a variable-length key. Once
is transmitted in a message’s clear text
more, 40-bit WEP cryptographic keys are
section. When used for cryptography, this
supported by the 802.11 standard. However,
24-bit string—which is used to start the key
some suppliers have used 128-bit or even
stream produced by the RC4 algorithm—is
104-bit keys in their products. The real
a comparatively small field. When an IV is
key used in the RC4 algorithm is 152 bits
reused, it generates identical key streams
for the 128-bit WEP key after the 24-bit
that protect data, and in a busy network,
IV is added. Noteworthy is the fact that
its short length ensures that these streams
some vendors generate keys based on a
will repeat after a short while. Furthermore,
user’s keystrokes; when done correctly,
the 802.11 standard is silent on the setting
with appropriate random processes, this
or modification of IVs, so different wireless
can yield a strong WEP key. As a result,
NICs from the same vendor might produce
the effective key size is usually smaller
identical IV sequences or some might even
than with other vendors who base WEP
employ a fixed IV. Because of this, hackers
keys on user-selected passwords. Security
are able to capture network activity, identify
flaws that allow malevolent users to breach
the key stream, and utilize it to crack the
WLAN security have been found by multiple
ciphertext.
teams of computer security experts. These
comprise statistical analysis-based passive The RC4 encryption key contains the
attacks to decrypt traffic, active attacks IV. Due to a flaw in the RC4 key schedule
to introduce new traffic from unapproved and the fact that an eavesdropper is aware
mobile stations (i.e., based on plain text of the first 24 bits of each packet key, a
that is known), ongoing attempts to decrypt successful analytic attack can recover the
traffic (i.e., dictionary-building attacks), and key after only a small amount of traffic
deceiving the access point attacks. Once has been intercepted and examined. The
sufficient traffic on a busy network has been attack script and open source code for this
analyzed, the dictionary-building attack can attack are accessible to the general public.
be launched. Cryptographic integrity is not protected
by WEP. To verify packet integrity and
Security problems with WEP include
acknowledge packets with the correct
the following:
checksum, the 802.11 MAC protocol employs
A well-known security flaw in wireless a noncryptographic Cyclic Redundancy
networks is the use of static WEP keys, Check (CRC).
 WIRELESS NETWORK SECURITY 283
It is risky and frequently results in vulnerabilities when noncryptographic checksums
and stream ciphers are combined, as is the case with WEP. By methodically altering
the packet, CRC sending it to the AP, and noting whether the packet is acknowledged,
the attacker can currently decrypt any packet. Because of the potential for interactions
with other protocol levels that could reveal cipher details, it is now deemed dangerous
to design encryption protocols without cryptographic integrity protection. These attacks
are frequently subtle.
Keep in mind that just one of the four issues mentioned above is dependent on a
cryptographic algorithmic flaw. Therefore, using a more powerful stream cipher would
not solve these issues. For instance, the third issue on the above list results from an
implementation flaw in the RC4 stream cipher that is made public by a badly thought
out protocol. Table 7.2 provides a summary of some of the issues with WEP and 802.11
WLAN security.
Table 7.2. Key Problems with Existing 802.11 Wireless LAN Security

Security Issue or Remarks

Vulnerability
1. Security features in Security features, albeit poor in some cases, are not
vendor products are enabled when shipped, and users do not enable
frequently not enabled. when installed. Bad security is generally better than
no security.
2. IVs are short (or 24- bit IVs cause the generated key stream to
static). repeat. Repetition allows easy decryption of data for
a moderately sophisticated adversary.
3. Cryptographic keys 40- bit keys are inadequate for any system. It is
are short. generally accepted that key sizes should be greater
than 80 bits in length. The longer the key, the less
likely a comprise is possible from a brute-force
attack.
4. Cryptographic keys Keys that are shared can compromise a system.
are shared. As the number of people sharing the key grows,
the security risks also grow. A fundamental tenant
of cryptography is that the security of a system is
largely dependent on the secrecy of the keys.
5. Cryptographic keys Cryptographic keys should be changed often to
cannot be updated prevent brute-force attacks.
automatically and
frequently.
6. RC4 has a weak The combination of revealing 24 key bits in the
key schedule and is IV and a weakness in the initial few bytes of the
inappropriately used in RC4 key stream leads to an efficient attack that
recovers the key. Most other applications of RC4 do
WEP. not expose the weaknesses of RC4 because they
do not reveal key bits and do not restart the key
schedule for every packet. This attack is available to
moderately sophisticated adversaries.
284 COMPUTER NETWORKS

7. Packet integrity is CRC32 and other linear block codes are inadequate
poor. for providing cryptographic integrity. Message
modification is possible. Linear codes are inadequate
for the protection against advertent attacks on data
integrity. Cryptographic protection is required to
prevent deliberate attacks. Use of noncryptographic
protocols often facilitates attacks against the
cryptography.
8. No user Only the device is authenticated. A device that is
authentication occurs. stolen can access the network.
9. Authentication Identity-based systems are highly vulnerable
is not enabled; particularly in a wireless system because signals
only simple SSID can be more easily intercepted.
identification occurs.
10. Device One-way challenge-response authentication is
authentication is subject to “man-inthe-middle” attacks. Mutual
simple shared-key authentication is required to provide verification
challenge-response. that users and the network are legitimate.
11.The client does not The client needs to authenticate the AP to ensure
authenticate the AP. that it is legitimate and prevent the introduction of
rogue APs.

7.2.4. Security Requirements and Threats

The WiFi, or 802.11 WLAN, market is growing and has a lot of momentum right now.
It appears that many organizations will implement 802.11 WLAN technology in the
upcoming years.

Figure 7.6. Taxonomy of Security Attacks (Source: Researchgate, https://www.researchgate.net/

profile/Shaurya-Gupta-10/publication/342685632/figure/fig2/AS:945114490875906@1602344262858/
Taxonomy-of-Security-Attacks.png).

Numerous establishments—such as shops, medical facilities, airports, and commercial

businesses—aspire to profit from “going wireless.” Nevertheless, despite enormous
success and growth, not everything about 802.11 WLANs has been favorable. Numerous
papers and reports detailing attacks on 802.11 wireless networks that put organizations at
risk for security have been published. To assist users and organizations in understanding
 WIRELESS NETWORK SECURITY 285
some of the attacks against WLANs, Figure 7.6 offers a general
taxonomy of security attacks.
Network security attacks are typically divided into passive and
active attacks. These two broad classes are then subdivided into
other types of attacks. All are defined below.
• Passive Attack: An attack in which an unauthorized party
gains access to an asset and does not modify its content (i.e.,
eavesdropping). Passive attacks can be either eavesdropping
or traffic analysis (sometimes called traffic flow analysis).
These two passive attacks are described below.
• Eavesdropping: The attacker monitors transmissions for
message content. An example of this attack is a person
listening into the transmissions on a LAN between two
workstations or tuning into transmissions between a wireless
handset and a base station.
• Traffic analysis: The attacker, in a more subtle way, gains
intelligence by monitoring the transmissions for patterns of Keyword
communication. A considerable amount of information is contained Denial-of-service
in the flow of messages between communicating parties. (DoS) attack is a
malicious attempt
• Active Attack: An attack whereby an unauthorized party to overwhelm an
makes modifications to a message, data stream, or file. It online service and
render it unusable.
is possible to detect this type of attack but it may not be
preventable. Active attacks may take the form of one of
four types (or a combination thereof): masquerading, replay,
message modification, and denial-of-service (DoS). These
attacks are defined below.
• Masquerading: The attacker impersonates an authorized
user and thereby gains certain unauthorized privileges.
• Replay: The attacker monitors transmissions (passive attack)
and retransmits messages as the legitimate user.
• Message modification: The attacker alters a legitimate
message by deleting, adding to, changing, or reordering it.
• Denial-of-service: The attacker prevents or prohibits the
normal use or management of communications facilities.
The risks associated with 802.11 are the result of one or more
of these attacks. The consequences of these attacks include, but
are not limited to, loss of proprietary information, legal and recovery
costs, tarnished image, and loss of network service.

7.2.4.1. Loss of Confidentiality

The quality of confidentiality is the ability to prevent information
from being accessed or revealed to unauthorized individuals,
286 COMPUTER NETWORKS

organizations, or processes. For most at least 100 MB of network packets have

companies, this is typically a fundamental been sniffed, the software computes the
security requirement. Confidentiality in a encryption keys while passively monitoring
wireless network poses a more complex the WLAN data transmissions. This much
security challenge due to the broadcast data can be collected in as little as three or
and radio nature of wireless technology. four hours on a highly saturated network;
Adversaries can access network resources in several days if traffic volume is low. For
without the need to physically connect a instance, after about 10 hours, a busy data
network cable. Additionally, the distance access point transmitting 3,000 bytes at
over which transmissions occur may be 11 Mbps will run out of 24-bit IV space.
uncontrollable. Consequently, traditional Data integrity and confidentiality could be
physical security measures become less readily jeopardized if an attacker manages to
effective. recover two cipher texts that were created
using the same key stream after ten hours.
Passively eavesdropping on native Once the packets on the network have been
802.11 wireless communications can pose a broadcast, monitoring poses an additional
significant risk to a company. An adversary risk to confidentiality breaches caused by
could potentially intercept and collect simple eavesdropping. An adversary can
sensitive data, including configuration observe traffic when an access point is
details, network IDs, passwords, and linked to a hub rather than a switch by
proprietary information. This risk arises from using a laptop in promiscuous mode. All
the possibility of an “insider” threat or the network traffic is typically broadcast by hubs
potential for 802.11 signals to extend beyond to all devices that are connected, making
the building. With the extended range the traffic open to unauthorized monitoring.
of 802.11 broadcasts, adversaries might Conversely, switches have the ability to be
be able to intercept transmissions from set up so that specific connected devices
nearby roads or parking lots. Exploiting this are not allowed to intercept broadcast traffic
vulnerability is relatively straightforward from other designated devices. Data meant
using a wireless network analyzer tool, or for both wired and wireless clients could be
sniffer, for two main reasons: (1) WLAN intercepted by a wireless device monitoring
technology’s confidentiality features are broadcast traffic, for instance, if a wireless
often left disabled, and (2) numerous security access point were linked to an Ethernet
vulnerabilities exist in 802.11 technology. hub. As a result, organizations connecting
These days, wireless packet analyzers to wireless access points ought to think
are easily accessible on the Internet thanks about utilizing switches rather than hubs.
to programs like WEPcrack and AirSnort. WLANs also run the risk of losing their
One of the earliest programs designed to confidentiality after an ongoing attack. As
automate network analysis was AirSnort. user names and passwords are sent over
Regretfully, it is also frequently employed a wireless connection, sniffing software—
to breach wireless networks. Vulnerabilities as previously mentioned—can obtain them,
in the key-scheduling algorithm included along with any other data that is traversing
in the original WEP standard and used the network. An adversary might be able
for RC4 implementation can be exploited to use a false identity to access a wired
by AirSnort. All that AirSnort needs to do network from an access point (AP). Once
this is a computer with a wireless network “on the network,” the hacker can use tools
card and the Linux operating system. Once
 WIRELESS NETWORK SECURITY 287
that are readily available to the public or that they have purchased
to scan the network. The malicious eavesdropper then gains access
to network resources and private company data by using the user
name, password, and IP address information.
Finally, there’s a security risk from rogue APs. A rogue AP could
be physically and covertly inserted into a closet, under a conference
room table, or into any other hidden space inside a building by a
malicious or careless user. Unauthorized users might then be able
to access the network through the rogue AP. The rogue AP can
successfully persuade wireless clients of its legitimacy and force
them to send traffic through it as long as it is situated close to
the WLAN users and is configured to look alike to them.
The wireless communication between wireless clients and an
authorized AP can be intercepted by a rogue AP. To intercept the
client traffic, it merely needs to be configured with a stronger
signal than the current AP. APs that are set up to permit access
without authorization can also be used by a malevolent user to
access the wireless network. It’s also crucial to remember that Remember
malevolent users don’t always deploy rogue access points. Rogue
Monitor your
APs are frequently set up by users who wish to utilize wireless
network for
technology without the IT department’s consent. Furthermore, intruders. You
because rogue APs a. should always
make sure you
have an eye on
7.2.4.2. Loss of Integrity what’s going
on, that you are
Issues with data integrity that arise in wired networks also occur tracking attack
in wireless networks. Integrity can be challenging to achieve trends. The more
because wireless and wired communications are commonly used you know about
what malicious
by organizations without sufficient cryptographic protection of data. security crackers
For instance, a hacker can jeopardize data integrity by altering or are trying to do
removing information from an email from a wireless system account. to your network,
the better the
This could be harmful to a company if crucial emails are shared job of defending
widely among the recipients. The lack of strong message integrity against them you
can do.
in the 802.11 standard’s current security features makes other active
attacks that jeopardize system integrity possible. A linear CRC is
all that the WEP-based integrity mechanism is. When cryptographic
checking techniques like message authentication codes and hashes
are not used, message modification attacks are possible.

7.2.4.3. Loss of Network Availability

A denial of service (DoS) attack, like jamming, is involved in a denial
of network availability. When a malevolent user intentionally emits
a signal from a wireless device to overpower authorized wireless
288 COMPUTER NETWORKS

signals, this is known as jamming. Emissions into company networks and systems by
from microwave ovens and cordless phones connecting to them via an untrusted
can also unintentionally cause jamming. network. Usually, users must gain access
Communication breaks down as a result of to resources that are classified as either
jamming because reputable wireless signals public or private by their organizations.
are unable to interact with the network. Transport Layer Security (TLS), the Internet
Another source of a DoS is non-malicious Engineering Task Force-standardized
users. For example, a user may download version of Secure Sockets Layer (SSL), is
large files and inadvertently monopolize a one application layer security protocol that
wireless signal, thereby preventing other agencies might want to think about utilizing
users from accessing the network. Agency to protect their public resources. In the
security policies should therefore restrict majority of agencies, though, this is not
the kinds and volumes of data that users required because the information is already
can download via wireless networks. available to the public. Agencies should
think about securing their connections to
private resources by utilizing a VPN solution,
Other Security Risks as this will help stop eavesdropping and
More users are looking for ways to connect unauthorized access.
remotely to the networks within their own Finally, social engineering and dumpster
organizations as a result of the widespread diving are issues as well, just like with any
use of wireless devices. The use of network. When preparing to implement a
untrusted, third-party networks is one such wireless network, an organization should
technique. For instance, wireless networks take network security into account from
are frequently available in conference every angle.
centers, enabling attendees to connect to
the Internet and then their own companies.
Airports, hotels, and even some coffee shops 7.2.5. Emerging Security
are starting to provide their patrons with Standards and Technologies
802.11-based publicly accessible wireless
networks. For an extra layer of security, Standards organizations have reacted to
these networks even offer VPN capabilities. the uproar over vulnerabilities in 802.11
WLANs, much like the security industry.
These untrusted public networks The IEEE and the Internet Engineering Task
present three main risks: 1) they are open to Force (IETF) are both active. Three distinct
the public and can therefore be accessed by projects are being worked on by the IEEE
anyone, including malevolent users; 2) they to enhance WLAN security. In the first,
act as a bridge to the user’s own network, as a long-term security solution, the IEEE
making it possible for anyone on the public 802.11 Task Group i (TGi) has suggested
network to attack or gain access to the substantial changes to the current IEEE
bridged network; and 3) they use high- 802.11 standard. On the basis of the
gain antennas to enhance reception and recently published Advanced Encryption
expand coverage area, making it easier for Standard (AES), the TGi is defining new
malevolent users to intercept user signals. ciphers. Although it will require new
Unless their organizations take action to hardware and protocol modifications, the
safeguard both their users and themselves, AES-based solution will offer a very reliable
users may introduce vulnerabilities solution in the future. TGi’s current design
 WIRELESS NETWORK SECURITY 289
requirements include the ability to detect
replay attacks and prevent forgeries, two Did you know?
of the most common issues with WEP. IEEE 802.11i enhances IEEE 802.11-1999 by providing
a Robust Security Network (RSN) with two new
The TGi’s temporary fix for WEP’s protocols: the four-way handshake and the group key
issues, WiFi Protected Access (WPA), is the handshake. These utilize the authentication services
second effort to enhance WLAN security. and port access control described in IEEE 802.1X to
To solve the issues without needing to establish and change the appropriate cryptographic
keys.
modify hardware, the team is developing
the Temporal Key Integrity Protocol (TKIP),
which will only require firmware and IEEE 802.1X can be used with a backend
software driver changes. The third initiative authentication server or fully implemented
from IEEE is the June 2001 approval of IEEE on the AP (by supporting one or more
802.1X-2001, a new standard that provides EAP methods within the AP). Diameter,
a generic framework for port-based network Kerberos, RADIUS, and other authentication
access control and key distribution. IEEE protocols are supported by the IEEE 802.1X
802.1X defines the encapsulation of EAP standard. RADIUS provides authentication,
(defined in RFC 2284) over IEEE 802 media, authorization, and accounting for Network
allowing mutual authentication between an Access Server (NAS) devices, such as dial-
AP and a station. The IEEE 802.11i draft up, xDSL, and 802.11; it is detailed in
standard specifies extra capabilities needed RFC 28652869 and RFC 3162. The 802.1X
for the safe implementation of IEEE 802.1X standard can be implemented using various
on 802.11 networks because IEEE 802.1X EAP types, such as EAP-TLS (defined in
was designed primarily for use with IEEE RFC 2716, supporting fast reconnect, mutual
802 LANs and not for use with WLANS. authentication, and key management via
Using an EAP method that supports mutual certificate authentication) for Ethernet
authentication, key management, and LANs and EAP-MD5 (defined in RFC 2284,
dictionary attack resistance are a few of supporting only one-way authentication
these requirements. Furthermore, 802.11i without key exchange). The IETF is
specifies a “four way” key management currently working on a new generation of
handshake that is used to confirm that EAP techniques that will address problems
the station has been authenticated to both with wireless authentication and key
the AP and, if one is present, a back-end management. Additional security features
authentication server. This handshake is like identity protection, secure ciphersuite
also used with the TKIP and AES ciphers. negotiation, cryptographic protection of
Therefore, it is crucial that IEEE 802.1X the EAP conversation, tunneling of other
implementations on 802.11 implement both EAP methods, etc., are supported by these
the fundamental IEEE 802.1X standard and methods. The reader is encouraged to visit
the IEEE 802.11i enhancements in order to the IEEE 802.11 standards website for the
provide sufficient security. most recent updates on each specification’s
status.

Did you get it?

1. What are the cost savings and operational efficiencies associated with WLAN deployments?
2. What emerging encryption protocols, authentication methods, and intrusion detection/prevention
systems are available for WLANs?
290 COMPUTER NETWORKS

7.3. WIRELESS PERSONAL

AREA NETWORKS
→Learning Objectives

• Gain an understanding of the fundamentals of Bluetooth technology

• Define the useful benefits of WPAN
• Explore strategies and best practices for securing Bluetooth devices and mitigating
potential security threats

Ad hoc networks are a relatively new wireless communication paradigm in which access
points and base stations are not part of a fixed infrastructure. Ad hoc networks rely on a
system of mobile routers connected by wireless links that allow devices to communicate
with one another, allowing devices to maintain random network configurations created
“on the fly.” In an ad hoc network, devices manage and exchange resources in addition
to controlling the network configuration. Ad hoc networks and peer-to-peer (P2P)
networks are comparable in that they both rely on decentralized networking, where
data is kept locally at the end user’s location as opposed to in a centralized database.
Ad hoc networks, on the other hand, are different from P2P networks in that the former
rely on the device hardware to request and share information, while the latter rely on
a routing mechanism to direct information queries. Ad hoc networks enable devices
to connect to wireless personal area networks (PANs) and use wireless applications,
like file sharing and address book synchronization. These networks can be expanded
to include network and Internet access by combining them with other technologies.
When connected to a Bluetooth network with another device that is capable of 802.11,
Bluetooth devices—which normally do not have access to network resources—can
establish a connection within the corporate network and access the Internet.

7.3.1. Bluetooth Technology

Nowadays, Bluetooth technology serves as the foundation for most ad hoc networks.
An open standard for short-range digital radio is called Bluetooth. It is marketed as a
low-profile, low-cost, low-power technology that offers an ad hoc means of establishing
small wireless networks. Bluetooth is regarded as a wireless PAN technology that
 WIRELESS NETWORK SECURITY 291
provides dependable and quick voice and data transmission. Untethered Bluetooth
devices function as a bridge to current networks and do away with the need for cables.
Bluetooth can be used to connect almost any device to any other device. An example
is the connection between a PDA and a mobile phone. The goal of Bluetooth is to
connect disparate devices (PDAs, cell phones, printers, faxes, etc.) together wirelessly
in a small environment such as an office or home. According to the leading proponents
of the technology, Bluetooth is a standard that will ultimately—
• Eliminate wires and cables between both stationary and mobile devices
• Facilitate both data and voice communications
• Offer the possibility of ad hoc networks and deliver synchronicity between
personal devices.
With some variations, Bluetooth is intended to function in the unlicensed ISM
(industrial, scientific, and medical applications) band, which is accessible in the majority
of the world. Table 7.3 provides a summary of Bluetooth’s features. While Bluetooth-
enabled devices can find each other automatically, users must take action to connect
with other devices and create networks. Like any other ad hoc network, Bluetooth
networks are created in an arbitrary and transient manner. The master-slave relationship
that is maintained between the network devices is one of the characteristics that sets
Bluetooth networks apart. A “piconet” is a master-slave network that consists of up to
eight Bluetooth devices. In a piconet, up to seven slave devices are directly connected
to the network that has one device designated as the master. The network is configured
and controlled by the master device, which also establishes the network’s hopping
scheme. Within a Bluetooth piconet, devices use the same frequency hopping pattern
and share the same channel. A slave in one network can function as the master for
another, even though only one device can serve as the master for each network.
Table 7.3. Key Characteristics of Bluetooth Technology

Characteristic Description
Physical Layer Frequency Hopping Spread Spectrum (FHSS).
Frequency Band 2.4 – 2.4835 GHz (ISM band).
Hop Frequency 1 ,600 hops/sec.
Data Rate 1 Mbps (raw). Higher bit rates are anticipated.
Data and Network Three modes of security (none, link-level, and service level), two levels
Security of device trust, and three levels of service security. Stream encryption for
confidentiality, challenge-response for authentication. PIN-derived keys and
limited management.
Operating Range About 10 meters (30 feet); can be extended to 100 meters.
Throughput Up to approximately 720 kbps.
Positive Aspects No wires and cables for many interfaces. Ability to penetrate walls and
other obstacles. Costs are decreasing with a $5 cost projected. Low power
and minimal hardware.
Negative Aspects Possibility for interference with other ISM band technologies.

Relatively low data rates. Signals leak outside desired boundaries.

292 COMPUTER NETWORKS

The dynamic network topologies of Bluetooth networks are

managed by mobile routers. Additionally, the routers manage the
data transfer between devices that can maintain a direct link. These
networks need to be dynamically reconfigured to accommodate the
devices’ erratic movements. Through the use of routing protocols,
Bluetooth is able to create and sustain these dynamic networks.
Similar to WLAN devices and other IEEE 802.11 compliant
devices, Bluetooth transceivers operate in the 2.4. GHz ISM band.
Frequency hopping (FH) spread spectrum system is used by Bluetooth
transceivers, which use Gaussian Frequency Shift Keying (GFSK)
modulation. The system hops 1,600 times per second across 79
frequencies in a quasi-random manner. A Bluetooth network can
have a maximum theoretical bandwidth of 1 Mbps. The networks’
inability to handle such data rates, however, is actually due to
communication overhead. It is anticipated that Bluetooth 2.0 will
offer up to 2 Mbps of maximum bandwidth.
Bluetooth networks are capable of supporting one channel
that transfers both synchronous speech and asynchronous data
Keyword simultaneously, or they can support one channel that transfers
Gaussian asynchronous data and up to three simultaneous synchronous speech
Frequency Shift channels. Circuit-switching and packet-switching technologies are
Keying or GFSK combined in Bluetooth. The benefit of Bluetooth packet switching
is an extension of
the FSK modulation is that it enables devices to route multiple information packets via
scheme where a single data path. The fact that this approach doesn’t use up all
the frequency of the resources on a data path makes it simpler for distant devices
the modulated
signal will not to keep data flowing through a scatter-net.
instantaneously
change at the
beginning of each 7.3.1.1. Brief History
symbol period of
the binary data. Ericsson Mobile Communication was the original architect of
Bluetooth, which bears the name of the Danish king Harald Bluetooth
from the tenth century. The Bluetooth SIG was established in 1998
by IBM, Intel, Nokia, and Toshiba and is responsible for overseeing
the specification. Initially, the SIG was established as a way to
track advancements in radio technology and the establishment of an
open, worldwide standard. The Bluetooth SIG is made up of more
than 2,000 organizations today, including prominent figures from the
computer and telecommunications sectors who are spearheading the
advancement and marketing of Bluetooth technology. When Bluetooth
was first developed, its main purpose was to replace cables in
wireless communication systems. To improve wireless connectivity,
SIG members intend to create a wide variety of consumer electronics
with Bluetooth capabilities. Cell phones, PDAs, laptop computers,
modems, cordless phones, pagers, PC cards, fax machines, and
 WIRELESS NETWORK SECURITY 293
printers are just a few of the many gadgets are symmetric, meaning that one device
that are expected. The IEEE 802.15 Personal can function both as a slave and a master.
Area Network (PAN) Working Group, which A fixed 48-bit unique device address (BD_
was established in the beginning of 1999, ADDR) is assigned to each radio. Piconets
has now standardized Bluetooth. There are ad hoc networks made up of two or
are many links to other websites on the more radio devices. A piconet’s units are
Bluetooth SIG website. all connected via the same channel. A
master device and one or more slaves are
present in every piconet. A piconet can
7.3.1.2. Frequency and Data Rates
contain up to seven active slaves at once.
Bluetooth was intended to function in As a result, a 3-bit active device address
the unlicensed 2.4 GHz–2.4835 GHz ISM uniquely identifies each active device in a
frequency band, just like the 802.11 WLAN piconet. It is possible for inactive slaves
standard. In order to mitigate interference in disconnected modes to stay inside the
caused by the many other technologies that piconet.
use this frequency range, Bluetooth employs
The only person who can start a
a frequency-hopping spread-spectrum (FHSS)
Bluetooth communication link is a master.
technology. By shifting frequency roughly
Nevertheless, the slave may ask for a
1,600 times per second, the FHSS scheme
master/slave switch to take control of the
makes use of 79 distinct radio channels. The
connection once it has been established. It
process is repeated continuously, using one
is forbidden for slaves to speak with one
channel for 625 microseconds, then hopping
another directly. All communication takes
in a pseudo-random order to another channel
place between the master and the slave.
for another 625 microsecond transmission.
Additionally, slaves in a piconet need to
As previously mentioned, because the
align their internal clocks and frequency
ISM band is free of license and accessible
hops with the master’s. A distinct frequency
everywhere, it has gained popularity for
hopping sequence is used by each piconet.
wireless communications.
Time Division Multiplexing was used by radio
Bluetooth technology can transmit data devices (TDM). In a piconet, the slaves may
at up to 1 Mbps in the ISM band, with a transmit on odd-numbered slots, while the
throughput of about 720 kbps. The data master device transmits on even-numbered
rates are still three to eight times faster slots. A scatternet is made up of several
than the average speed of parallel and serial piconets with overlapping coverage areas.
ports, respectively, even though they are While slaves may participate in multiple
lower than those of 802.11 wireless LANs. piconets on a time-division multiplex basis,
For many of the applications that Bluetooth each piconet may have a single master. A
was designed for, this rate is fast enough. device can function as a slave in multiple
Furthermore, even faster data rates are piconets or as a master in one and a slave
expected to become available in the future. in another.

7.3.1.3. Bluetooth Architecture and 7.3.1.4. Range

Components
Three distinct classes of power management
A slave radio can communicate with a are offered by Bluetooth. The highest
master radio via Bluetooth. Bluetooth radios power devices, known as Class 1, have an
294 COMPUTER NETWORKS

operating range of up to 100 meters (m) and run at 100 milliwatts (mW). The lowest
power Class 3 devices operate at 1 mW and have an operating range of 1/10 meter to
10 meters. Class 2 devices operate at 2.5 mW and have an operating range of up to 10
meters. Table 7.4 provides an overview of these three operating power levels.
Table 7.4. Device Classes of Power Management

Type Power Power Level Operating Range

Class 1 High 100 mW (20 Up to 100 meters (300
Devices dBm) feet)
Class 2 Medium 2.5 mW (4 dBm) Up to 10 meters (30
Devices feet)
Class 3 Low 1 mW (0 dBm) 0.1–10 meters (less
Devices than 30 feet )

In Figure 7.7, the three Bluetooth ranges are shown. The shortest range, as
demonstrated, might be useful for tasks like changing out cables (e.g., mouse or
keyboard), syncing files, or exchanging business cards. Like with the data rates, it is
expected that even longer ranges will be attained in the future. The high-powered range
can travel up to 100 m, or roughly 300 feet.

Figure 7.7. Bluetooth Operating Range (Source: Geeksforgeeks, https://media.geeksforgeeks.org/

wp-content/uploads/20190322083717/bluetootharchitecture.jpg).

7.3.2. Benefits
Users can benefit from Bluetooth in five main ways. Bluetooth is becoming increasingly
appealing due to its ad hoc untethered communication method, which can lead to lower
costs and more efficiency. Both home users and business users in large organizations
find the efficiencies and cost savings appealing.
Benefits of Bluetooth include—
• Cable replacement: For a range of interconnections, Bluetooth technology takes
the place of cables. These consist of those of auxiliary equipment (i.e., USB at
12 Mbps (USB 1.1) up to 480 Mbps (USB 2.0); printers and modems, typically
 WIRELESS NETWORK SECURITY 295
at 4 Mbps; wireless headsets and would turn on a cell phone or any
microphones that interface with other gadget that could establish
PCs or mobile phones; mouse and a network connection. After the
keyboard computer connections. person logs on, the laptop would
• Ease of file sharing: File sharing then automatically start a network
between Bluetooth-enabled devices join by using the phone as a modem
is made possible via Bluetooth. For and send the emails.
instance, laptops that are Bluetooth • Internet connectivity: Many
compatible during a meeting allow different kinds of devices and apps
participants to share files. Another support Bluetooth. Mobile phones,
example is the use of a mobile PDAs, laptops, desktop computers,
phone with Bluetooth compatibility and fixed telephones are a few
as a wireless modem for laptops. of these gadgets. When these
The laptop communicates with the technologies and devices work
mobile phone through Bluetooth, together to leverage each other’s
and the phone connects to a capabilities, internet connectivity
network. This allows the laptop to is made possible. For instance,
have full networking functionality a laptop can ask a smartphone
without requiring an electrical to establish a dial-up connection
interface to connect to the mobile via Bluetooth, and the laptop can
device. use that connection to access the
• Wireless synchronization: Internet.
Automatic wireless synchronization Office equipment is anticipated to include
with other Bluetooth-capable Bluetooth built-in (e.g., laptops, printers,
devices is made possible by fax machines, and PCs), communication
Bluetooth. For instance, PDAs, appliances (e.g., pagers, headsets, cell
laptops, mobile phones, and other phones, and handsets), as well as household
devices can synchronize the private appliances (e.g., microwave ovens, DVD
data found in address books and players, cameras, and refrigerators). In
date books. addition, Bluetooth finds use in smart
• Automated wireless applications: homes, banking, electronic payment systems,
Automatic wireless application vending machines, wireless conference
functions are suppor ted by rooms and offices, in-car communications,
Bluetooth. In contrast to local and parking.
synchronization, automatic wireless
applications communicate with the Did you know?
LAN and Internet. A person working The development of the “short-link” radio technology,
offline on emails, for instance, later named Bluetooth, was initiated in 1989 by Nils
might be traveling or otherwise Rydbeck, CTO at Ericsson Mobile in Lund, Sweden.
not within their usual service The purpose was to develop wireless headsets,
according to two inventions by Johan Ullman, SE
area. After returning to a service
8902098-6, issued 1989-06-12 and SE 9202239,
location, the person will email issued 1992-07-24.
the files that are queued in the
laptop’s inbox (i.e., upon landing),
296 COMPUTER NETWORKS

7.3.3. Security Requirements instance, can be used as bugging tools by

malevolent users. While successful attacks
and Threats
on PCs using Back Orifice and Netbus have
Bluetooth provides a number of advantages been reported, such attacks have not been
and benefits. Before implementing the documented because Bluetooth is not yet
technology, organizations must, however, not widely used commercially. A malevolent
only address the security risks associated user may be able to access other Bluetooth
with Bluetooth but also assess the devices and networks with little to no
vulnerabilities of the devices they allow to security if they have a program like Back
connect to Bluetooth networks. Organizations Orifice installed on a device connected to
must specifically focus on security issues the Bluetooth network. It is possible to
related to network availability, data use these same programs against Bluetooth
integrity, and confidentiality. Additionally, networks and devices. Because the system
Bluetooth devices are more susceptible to authenticates the devices rather than the
unintentional security breaches as they users, Bluetooth devices are even more
are typically managed by users who may vulnerable. This means that a compromised
have lower security awareness compared device has the ability to access the network
to administrators. and compromise the network itself as well
as the devices connected to it..
7.3.3.1. Confidentiality Loss Bluetooth networks are vulnerable
For a comprehensive overview of security to unauthorized remote users. Users at
attacks, refer to Figure 7.7 in the 802.11 a distance are not always bound by the
wireless section. This will help in same security regulations as those in
understanding some of the attacks that target person. Whether they are traveling or at
Bluetooth. Compromised Bluetooth devices home, they routinely use insecure links.
pose the primary threat to confidentiality. They send user IDs and passwords during
When a component of a Bluetooth piconet the connection process, which a malevolent
is compromised (e.g., in the possession of user can intercept with a network sniffer.
an unauthorized individual), sensitive data A malicious user does not need to be
may be accessed by the malicious party. physically close to the user in order to
Moreover, the compromised device may intercept traffic because there is no secure
retain information or network privileges perimeter as is usually present in an office
that could jeopardize the broader network. setting. All of the devices in that Bluetooth
In such cases, the compromised device network are open to attacks as soon as the
may request data as part of a deliberate device or link is compromised. For instance,
network attack in addition to receiving a malicious user can request and receive
regular traffic. The fact that Bluetooth sensitive data from a compromised device,
networks require device authentication and they can monitor data traffic via a
rather than user authentication for resource compromised link. A compromised device
access is a distinguishing feature of this may also be used to access the network
compromise. Once authenticated, the device if the malicious user manages to learn the
does not need further authentication and password and user IDs for the targeted
gains immediate access to resources. The network. In this case, it takes several
security of Bluetooth devices is inherently security lapses before a malevolent user
compromised. Wireless microphones, for is able to access the network. Mitigating
 WIRELESS NETWORK SECURITY 297
the risk of such an attack would involve Device A or Device B won’t ever be aware
implementing additional security layers on that their data is being monitored, which
top of Bluetooth and Bluetooth secure links. poses the greatest risk.
An additional risk to Bluetooth devices that
use unit keys—generally the more basic
“dumb” devices—comes from the man-in-
the-middle attack. The security encryption
key that a network device (Device A) uses
to monitor traffic between itself and another
network device (Device B) is obtained by
the man-in-the-middle (Device C) in this
attack. Device A only needs to share its
unit key—a static key that is specific to
each device—separately with Devices C
and B in order for the attack to succeed.
There may be no connection at all between Figure 7.8. Man-in-the-Middle Attack Scenarios
Devices A and B and Devices A and C, (Source: Wallarm, https://uploads-ssl.webflow.
and there may even be a difference in the com/5ff66329429d880392f6cba2/605cab5ff8f386
degree of secrecy. But once Device C is ea033ae16c_Man-in-the-Middle)%20Attack.jpg).
aware of the unit key, it can determine the
Network administrators cannot see
encryption key and surreptitiously monitor
Bluetooth devices, and there is currently
communication between Devices A and B by
no software available to monitor these
using a fictitious device address. There is no
intrusions. In an ad hoc network, it
expensive or specialized equipment needed
doesn’t really matter if users from various
for the man-in-the-middle attack. The attack
organizations enforce different security
can be carried out by a skilled malevolent
policies. Every device connected to the ad
user who has access to the unit key and
hoc network is vulnerable to each other’s
can impersonate a Bluetooth address to
security flaws. Since network administrators
generate the encryption key. Such attacks
are unlikely to be in charge of Bluetooth
rely on prior knowledge about the Bluetooth
devices, users should be mindful of the
devices that are being targeted. However,
security risks when using them in settings
this does not always mean that malicious
where sensitive data is processed. While
users cannot use rand.
agencies should take privacy concerns into
The attack is depicted in Figure 7.8. account when implementing Bluetooth
Device A, a reliable PDA, exchanges technologies, privacy violations are not a
confidential data with Device B, a reliable direct threat to security. Every Bluetooth
laptop. Device A connects to Device C, an gadget has a distinct address (BD_ADDR),
untrusted PDA, during the connection with which is utilized to record every device’s
Device B in order to exchange personal involvement in the network. In order to
contacts from A’s PDA address book. Using guarantee device authentication (i.e.,
Device A’s unit key and a fictitious address, because the address that is logged belongs
Device C can monitor traffic between to the device and not to the user, an
Devices A and B after connecting to Device individual can refuse to participate in the
A. This makes Device C the man-in-the- network since we lack evidence of who
middle. It’s possible that the owner(s) of was using the device. But it also enables
298 COMPUTER NETWORKS

organizations to keep an eye on and record DoS attacks prevent authorized users from
what. accessing network applications and system
resources. Aside from the standard DoS
attacks (e.g., Signal jamming attacks (such
7.3.3.2. Loss of Integrity
as flooding techniques) can also target
When users’ or organizations’ data is Bluetooth devices, LANs, and Internet
tainted, there are violations of integrity. The services. Bluetooth devices are susceptible to
immediate result is a compromised network, interference because they share bandwidth
much like in the case of a confidentiality with cordless phones, microwave ovens, and
or disclosure threat. Beyond this, though, other wireless networks. Malicious users
are threats to network integrity that have the ability to obstruct the information
include information being added, altered, flow (i.e., employing devices that broadcast
or deleted and then transmitted through in the 2 GHz ISM band to skew the network’s
the network without the user’s or network information and interfere with the routing
administrator’s knowledge. Data on user protocol. Ad hoc network devices are not
devices and files on the network are both able to negotiate the dynamic topologies
susceptible to corruption. For instance, a of the network when the routing protocol
malevolent user may use a PDA or other is disrupted. Jamming may occur more
untrusted device to access the address book frequently for remote users than for on-
of a different PDA or laptop. But unlike a site users. The same interference that users
disclosure threat, where information would face in the office must also be dealt with
only be monitored, the malicious user by remote users. Additionally, because the
changes the contact details without the remote environment is uncontrolled, it is
owner’s knowledge or even erases them more probable that distant devices will be
entirely. Such attacks may cause the agency near other devices (e.g., other ISM band and
(or user) to lose faith in its data and system Bluetooth devices) that are purposefully or.
if they go unnoticed. It is recommended that
An attack due to battery exhaustion is
users confirm that their Bluetooth product
another risk connected to ad hoc devices.
prohibits automatic data synchronization in
This attack uses a device’s battery to try
order to stop any information from being
and disable it. The device receives frequent
changed without the user’s consent.
requests from a malevolent user requesting
either the creation of a network or data
7.3.3.3. Loss of Availability transfers (presuming the user is a member
of the network topology). This kind of attack
Network availability and “usability upon doesn’t jeopardize network security, but in
demand” for authorized users and devices the end, it keeps the user from accessing
are lost as a result of DoS and DDoS attacks. the network since it breaks the device.

Did you get it?

1. What are the architecture and components of Bluetooth technology?
2. What are the implications of a loss of confidentiality in communications?
 WIRELESS NETWORK SECURITY 299

ROLE MODEL

Norman Manuel Abramson: An American engineer and computer scientist, most known
for developing the ALOHA net system for wireless computer communication.

BIOGRAPHY
Norman Abramson, the leader of a group of scientists and engineers who pioneered
the development of wireless computer networks, passed away on December 1 at his
residence in San Francisco. He was 88.
The cause of his passing was skin cancer that had spread to his lungs, as reported
by his son, Mark.
Professor Abramson’s project at the University of Hawaii was initially intended
to transmit data to schools on the distant Hawaiian islands through a radio channel.
However, the solution he and his team developed in the late 1960s and early 1970s
would prove to be widely applicable; some of their technology is still utilized in today’s
smartphones, satellites, and home WiFi networks.
The technology they invented enabled numerous digital devices to exchange data
over a shared radio channel. It was a straightforward approach that did not necessitate
intricate scheduling of when each data packet would be transmitted. If a data packet was
not received, it was simply retransmitted. This method deviated from telecommunications
practices at the time, but it was effective.
“It was an incredibly bold idea, truly innovative engineering,” remarked Vinton
Cerf, a computer scientist at Google and the co-author, alongside Robert Kahn, of the
technical standards for connecting computer networks on the internet.
The wireless network in Hawaii, which commenced operations in 1971, was named
ALOHA net, incorporating the Hawaiian greeting or farewell. It served as a smaller,
wireless counterpart to the more renowned ARPA net, the precursor to the internet, which
enabled university researchers to communicate over a network and send messages via
landlines. The ARPA net was overseen by the Pentagon’s Advanced Research Projects
Agency, which also supported the ALOHA net.
300 COMPUTER NETWORKS

“The early wireless work in Hawaii is vastly underappreciated,” said Marc Weber,
an internet historian at the Computer History Museum in Mountain View, Calif. “Every
modern form of wireless data networking, from WiFi to your cellphone, goes back to
the ALOHA net.”
Professor Abramson has been called the father of wireless networking. But it was a
shared paternity. The project included graduate students and several faculty members,
notably Frank Kuo, a former Bell Labs scientist who came to the University of Hawaii
in 1966, the same year Professor Abramson arrived.
His deepest expertise was in communication theory, the subject of his Ph.D. thesis
at Stanford University. The fundamental design ideas behind ALOHA net were his. In
a 2018 oral history interview for the Computer History Museum, Professor Kuo recalled,
“Norm was the theory and I was the implementer, and so we worked together pretty
well.”
ALOHAnet owed a lot to surfing. Professor Abramson was presenting a paper at
an academic conference in Tokyo in the days when flights from San Francisco to Tokyo
had to stop midway in Honolulu. Professor Abramson, who was raised in Boston, had
not been to Hawaii before and decided to spend a few days there on the way home.
He rented a surfboard. “I got on, I learned how to surf, and I said, Boy, I could
stand some of this,” he recalled in 2013 in an oral history interview with the Computer
History Museum.
Within a year, after the University of Hawaii offered him a tenured professorship,
he and his family moved to Hawaii. “My father was really wrapped up in his work, but
he surfed nearly every day,” Mark Abramson said.
That the ALOHAnet technology became so widely used was partly because Professor
Abramson and his team had shared it freely and welcomed other scientists to Hawaii.
“We had done no patenting, and ALOHA was published in scientific papers,” putting
their work in the public domain, Professor Abramson said in the oral history, adding:
“And that was fine with me. I was too busy surfing to worry about that sort of thing.”
Norman Manuel Abramson was born in Boston on April 1, 1932, to Edward and
Esther Abramson. His father was a commercial photographer, his mother a homemaker.
Norman and his sister, Harriet, grew up in the Dorchester neighborhood, home to mostly
Jewish immigrants, like his parents, at the time. His father was from Lithuania, his
mother from Ukraine.
Norman was educated in Boston’s public schools, at the elite Boston Latin School
and the English High School, where he excelled in math and science. He went on to
Harvard University, where he took a course taught by Howard Aiken, a mathematician
and early pioneer in computing. It was a computer course long before the discipline of
computer science existed, and he enjoyed his first taste of programming.
 WIRELESS NETWORK SECURITY 301

Professor Abramson majored in physics at Harvard, then earned a master’s degree

in physics from the University of California, Los Angeles, and his doctorate in electrical
engineering from Stanford, in 1958. He briefly worked in industry and had postdoctoral
teaching stints before he went to Hawaii. He retired from the University of Hawaii in
1994.
In addition to his son, Mark, he is survived by his wife, Joan Abramson; his sister,
Harriet Schannon; and three grandchildren. His daughter, Carin Wethington, died in 2014.
Some of the data-networking techniques developed by Professor Abramson and his
Hawaii team proved valuable not only in wireless communications but also in wired
networks. One heir to his work was Robert Metcalfe, who in 1973 was a young computer
scientist working at Xerox PARC, a Silicon Valley research laboratory that had become
a fount of personal computer innovations.
Mr. Metcalfe was working on how to enable personal computers to share data
over wired office networks. He had read a 1970 paper, written by Professor Abramson,
describing ALOHAnet’s method for transmitting and resending data over a network.
“Norm kindly invited me to spend a month with him at the University of Hawaii to
study ALOHAnet,” Mr. Metcalfe recalled in an email.
Mr. Metcalfe and his colleagues at Xerox PARC adopted and tweaked the ALOHAnet
technology in creating Ethernet office networking. Later, Mr. Metcalfe founded an Ethernet
company, 3Com, which thrived as the personal computer industry grew.
“Norm, thank you,” Mr. Metcalfe concluded in his email. “Aloha!””
302 COMPUTER NETWORKS

SUMMARY
• A wireless network is one that allows computers and other devices to communicate
using radio signal frequency. It’s also referred to as a WLAN or WiFi network
sometimes.
• A wireless LAN, or wireless local area network, consists of two or more computers
linked to the same radio frequency. In contrast, a wired LAN requires each
computer to be physically connected to the server’s network switch or hub via
an Ethernet cable.
• Physical Ethernet cables are used to connect the desktop server to the DSL
modem, switch, and router. But the clients don’t need any cabling.
• Since their beginning, wireless networks have been categorized into ad hoc
networks and infrastructure networks, which encompass LANs. The term “ad
hoc” in Latin means “for this purpose.” Ad hoc networks are designed for
specific purposes, often created quickly and for a temporary use.
• A minor backlash appeared in popular culture due to the rapid rise to fame of
the personal digital assistant and its early associations as a status symbol for the
business class.
• A wireless local area network, or WLAN, consists of two or more computers
connected to the same radio frequency. In contrast, a wired LAN has computers
physically connected to the server’s network switch or hub via Ethernet cables.
• A WLAN comprises a wireless station and an access point. A laptop or notebook
PC with a wireless network interface card (NIC) is commonly known as a station
or client.

MULTIPLE CHOICE QUESTIONS

1. Which of the following is a fundamental security protocol used to secure wireless
networks?
a. WEP
b. WPA2
c. WEP2
d. WPA
2. What does WEP stand for in the context of wireless network security?
a. Wired Equivalent Privacy
b. Wireless Encryption Protocol
c. Wi-Fi Encryption Protection
d. Wireless Enhanced Privacy
3. Which of the following authentication methods is considered the most secure for
wireless networks?
a. WPA-PSK
 WIRELESS NETWORK SECURITY 303
b. WEP
c. WPA2-Enterprise
d. MAC Filtering
4. What does SSID stand for in the context of wireless networks?
a. Secure Server Identifier
b. Service Set Identifier
c. Secure Signal Detection
d. System Security Identifier
5. Which of the following is a common method used to prevent unauthorized access
to a wireless network?
a. Enabling WPS (Wi-Fi Protected Setup)
b. Broadcasting SSID
c. Disabling encryption
d. MAC address filtering

REVIEW QUESTIONS
1. What are Ad Hoc Networks?
2. How are Personal Digital Assistants (PDAs) relevant in wireless networking?
3. What are some examples of Emerging Wireless Technologies?
4. What components constitute Wireless LANs?
5. How is the security of 802.11 Wireless LANs ensured?

Answer to Multiple Choice Questions

1. (b) 2. (a) 3. (c) 4. (b) 5. (d)

REFERENCES
1. Andy, D., (2002). Emerging Technology: Wireless LAN Standards. NetworkMagazine.
com. http://networkmagazine.com/article/NMG20020206S0006 (accessed on 05
April 2024).
2. Bajwa, M. I., (2004) Wireless Network Security. Directed Study Project, Strayer
University, Herndon.
3. Vacca, J. R. (2013). Computer and Information Security Handbook (2nd ed.). Morgan
Kaufmann.
4. Rappaport, T. S. (2002). Wireless Communications: Principles and Practice (2nd
ed.). Prentice Hall.
5. Stallings, W. (2017). Wireless Communications & Networks (2nd ed.). Pearson.
304 COMPUTER NETWORKS

6. Russell, D. L., Ganguly, A., & Liscano, R. (2016). Wireless Security: Models, Threats,
and Solutions. CRC Press.
7. Pirzada, A. A., & McDonald, C. (2007). Network Security: A Beginner’s Guide (2nd
ed.). McGraw-Hill.
Index

A Broadcast mechanism 102

Burned-in addresses (BIAs) 145
Access control lists (ACLs) 98
Business model 3
Adaptability 143
Addressing scheme 98, 105 C
Address resolution functions 146
Central coordination 222
Advanced Research Project Agency Network
Centralized management 98
(ARPANET) 148
Civil Aviation Authorities 3
Advanced Television System Committee
Client requests 224, 227, 234, 242
(ATSC) 51
Client-server based system's performance 226
Affiliations 193
Client-server model 221, 224, 225, 227, 229,
Application layer 52
230, 231, 232, 236, 254
Application tier 234, 235
Cloud computing 49
Arbitrary groupings 147, 178
Cluster boundary information 194
Artificial intelligence 183, 217
Cluster head 185, 186, 187, 188, 194, 198, 199,
Asynchronous Transfer Mode (ATM) 101, 114
200, 204, 205
B Cluster maintenance 186
Cluster networks 181, 183, 217
Backbone Networks (BBN) 5, 38
Communication medium 102, 122
Bandwidth 98, 114, 117, 121, 127, 128, 129, 130
Communication network 184, 206, 207
Bandwidth management 2
Communications subsystem 143
Battery storage room 2
Communications system 99
Berkeley Software Distribution (BSD) 143
Computational fluid dynamics 183, 217
Biomedical simulations 183, 217
Computational server 229
Blockchain technology 221
Computer circuit 65
Blu-ray players 106
Computer data storage logic 233
Broadcast environments 185, 218
Computer network 4, 5, 6, 7, 26, 28, 29, 32, 33,
Broadcasting range 262, 270
34, 37, 44, 45, 46
306 COMPUTER NETWORKS

Computer networking technology 139 Elasticity 182

Computer science 7, 8, 10, 26, 40, 42 Electronic components 64
Computer transmitting data 7 Employee sabotage 259
Computing environment 100, 102 Encryption 53
Computing hardware 5 Enterprise Private Networks 5, 34, 35, 39
Confidentiality 257, 265, 268, 270, 274, 275, Equipment storage 2
280, 285, 286, 291, 296, 298 Ethernet 51, 57, 84
Configuration errors 184, 218 Ethernet cable 4, 30, 46, 105
Connectivity 184, 185, 186, 187, 188, 189, Exchange packets 147
195, 197, 209, 212, 216, 218 Extensibility 225
Consolidation 182
F
Consulting 146
Content delivery networks (CDNs) 222 Fiber Distributed Data Interchange (FDDI)
Continuous communication protocols 141 101
Conventional multiplexers 67 File transfer 107
Cost-effectiveness 181 File Transfer, Access, and Management
(FTAM) 53
D
Financial modeling 183, 217
Database services 224 Fixed telephony 2
Database Tier 234, 236 Foreign espionage 259
Data breaches 257 Framework 49, 87
Data compression 54 Frequency assignment 189
Data encryption 106
G
Data-link addresses 144, 145, 178
Data link layer 52 Gigabit Ethernet 51
Data-link layer addresses 143, 144, 145 Global Area Networks (GAN) 5, 39
Data management 231 Google File System (GFS) 222
Data redundancy 103
Data storage devices 99 H
Denial of service (DoS) 266 Hadoop Distributed File System (HDFS) 222
Dependability 266 Hardware addresses 144, 161, 178
Desktop computers 4, 101, 107 Hardware cabling 106
Desktop server 261, 269, 302 Hardware resources 101, 136
Destination node 189 Hexadecimal digits 144, 145, 146
Dialog Control 54 Hierarchical address space 147
Digital applications 64 Hierarchical routing 184, 191, 192, 193, 194,
Digital data 58, 62, 64, 65, 94, 99 195, 196, 197, 200, 218
Digital multiplexer 64 High-Definition Television (HDTV) 51
Distributed computing architecture 223 High-performance computing (HPC) 183,
Distributed database 228, 249 217, 218
Distributed network systems 221 Home Area Networks (HAN) 5, 31, 39
Distributed operating system 237 Hub 101, 104, 108, 109, 120, 121
E I
 INDEX 307
Identifier-based clustering 186 145
Impromptu networks 262 Metal shelf structure 2
Inclination 3 Metropolitan Area Networks (MAN) 5, 33
Information age 101, 136 Mobile devices 264, 291
Information sharing 101, 108, 114, 136 Mobile networks 184, 218
Information technology resource 266 Mobile operating systems 264
Infrared (IR) technology 260 Mobile wireless network 188
Integrated development programs 2 Modularity 225
Integrity 257, 265, 268, 270, 274, 275, 278, Multifunctional link 4
281, 282, 283, 284, 286, 287, 296, 298 Multiple-Input Multiple-Output (MIMO) 261
Interface card 102 Multiplexer 64, 67
International Standards Organization 52, 94
N
Internet café 3
Internet connection 261, 262, 263, 269 Name server 226, 237, 239, 240
Internet Control and Configuration Board National Television System Committee
(ICCB) 142 (NTSC) 51
Internet Message Processors (IMP) 142 Natural cooling techniques 3
Internet Protocol (IP) 7 Near-Term Digital Radio Network (NTDR)
Internetwork addresses 143 184
Internetworking 139, 140, 141, 142, 144, 145, Network availability 266, 298
147, 148, 149, 153, 178, 179, 181 Network Control Program (NCP) 142
Internetwork Packet Exchange (IPX) 146 Network expansion 143
Interoperability 49, 51, 53, 83, 87, 100 Networking algorithms 189
Intra-cluster transmission 186 Networking communications 7
Intrusion detection systems (IDS) 258 Networking software 51, 94
Intrusion prevention systems (IPS) 258 Networking suppliers 52
Network interface 144, 147, 161
J
Network interface cards (NICs) 261, 269
Jeopardize 265, 287, 296, 298 Network Layer 50, 56, 94
Network-layer addresses 143, 147
L
Network loops 98
Landmark routing 193 Network management 143, 148
Latency 183, 195, 205, 217, 218, 219 Network management software 223
Laterite mud bricks 3 Network Near-Term Digital Radio (NTDR)
Link-cluster architecture 184 networking 187, 218
Link-state data 192, 193, 195, 197 Network Operating Center 2
Link state information 188 Network product 140
Local area network (LAN) 4, 30, 32 Network protocols 7, 8
luster membership 186 Network security 258, 259, 274, 277, 288,
298, 302
M Network segments 139
Marketability 52 Network standards 51
Media Access Control (MAC) addresses 143, Network topology 97
308 COMPUTER NETWORKS

Nnetwork support divided subnets 189 R

Numerical address 102
Radio signal frequency 260, 302
O Random-access memory (RAM) 145
Read-only memory (ROM) 145
Office setting 101
Reliability 97, 113, 121, 229, 236
One-to-many communication pattern 228,
Reliable services 227
229, 232
Remote computer 53, 54
Online businesses 49
Remote direct memory access (RDMA) 183,
Open Environment Corporation (OEC) 233
217
Open Systems Interconnection (OSI) 4, 6, 48
Repeater 104
Operating systems (OS) 264
Request-response pattern 230
Output signal 64
Resource management 231, 245
P Resource utilization 50, 221, 244, 246
Risk analysis 183, 217
Parallel computing 183, 217
Rural Wireless 2
Parallel execution 227, 228, 229
Parallel processing 222, 228 S
Peer-to-peer (P2P) networks 222
Scalability 97, 126, 134
Peripheral devices 4
Scanners 99
Personal Area Networks (PAN) 5, 30
Seamless communication 97
Personal computers 99, 100, 107, 109
Security 97, 98, 106, 107, 109, 117, 121, 122,
Personal digital assistant (PDA) 263
126, 127, 134
Personal information manager 264
Security configurations 258
Physical Ethernet cables 261, 269, 302
Security risks 265
Physical infrastructure 50
Self-organizing controlled structures 184
Physical layer 52, 58
Sequence numbers 72, 74
Physical network association 144, 147
Server 103, 104, 105, 106, 107, 108, 113, 118,
Physical space 99
129, 134, 136
Physical subnet 189, 190
Server room 2, 3
Power backup systems 2
Serviceable performance 101
Power Holding Company of Nigeria (PHCN)
Service access point (SAP) 79, 80
2
Service discovery 238
Presentation layer 52, 54
Session layer 52, 55
Presentation Tier 234, 236
Shared database management 53
Printers 99, 100, 102, 104, 105, 106, 107
Shortest path-first (SPF) algorithm 188
Print services 104, 107
Signal light 3
Programming teams 235
Smartphones 105
Prominent illustration 141
Software applications 51
Proof of Stake (PoS) 222
Software-defined networking (SDN) 51
Proof of Work (PoW) 222
Software entity 225
Protocol standards 100
Solar energy 2
Public transmission system 99
Source computer 7
Space exploration 183, 217
 INDEX 309
Spanning Tree Protocol (STP) 98 Unmanaged networks 148
Special Publication (SP) 265 Urgent Pointer (URG) 72
Standard Internet Checksum algorithm 80 User Datagram Protocol (UDP) 149, 151, 158
Standardization 102
V
Standard logic gates 64
Standard reference model 139, 140 Vector routing information 193
Sun Workstation 104 Vendor's products 52
SURAN (Survivable Adaptive Networks) 193 Virtual communication 225
Synchronization 266, 290, 295, 298 Virtualization technology 49
Synchronous transmission 99 Virtual subnets 188, 189
Systems Network Architecture (SNA) 4
Systems support 147
W
Web server 235, 242, 243
T
Wide-area networks (WANs) 4
Telephone Network 65 Wireless cards 261
Trajectory analysis 183, 217 Wireless clients 261, 278, 281, 286, 287
Translation task 53 Wireless communications 2
Transmission 7, 10, 18, 29, 34, 45 Wireless hotspot 261
Transmission Control Protocol (TCP) 141, Wireless local area network (WLAN) 261
149, 151 Wireless networking standards 51
Transmission media 104 Wireless network security 257
Transmitter 263, 273, 276 Wireless sensor networks (WSNs) 198
Transparent data distribution 103 Wireless transmission 185
Transport layer 52, 55 Workgroup style networks 262
Workstations 100, 102, 103, 104, 109, 136
U
World Wide Web (WWW) 148, 153
ultiple physical network connections 144
X
Xerox Network Systems (XNS) 146