www.gradamic.

com

SECTION A: Short/Brief Questions

1. What are the applications of Computer Networks?

Computer networks have many important uses in our daily lives:

• Communication: Email, instant messaging, video calls (like Zoom or Skype)

• Resource sharing: Sharing printers, files, and storage between multiple computers

• Information access: The internet lets us access websites, news, and online databases

• Entertainment: Streaming movies (Netflix), online gaming, and social media

• E-commerce: Online shopping (Amazon), banking, and payment systems

• Remote work: Accessing office systems from home or anywhere

• Cloud services: Storing files online (Google Drive, Dropbox) and using web apps

2. Explain differences between point-to-point and point-to-multipoint.

• Point-to-point: A direct connection between exactly two devices, like a wire between two
computers. It's private and dedicated.

• Point-to-multipoint: One device connects to multiple devices, like a Wi-Fi router connecting
to many phones/laptops. It's shared and can broadcast to many.

3. Define fixed and variable size framing.

• Fixed size framing: Data is divided into same-size pieces (like ATM cells always being 53
bytes). Easier to handle but may waste space.

• Variable size framing: Data pieces can be different sizes (like Ethernet frames from 64 to
1518 bytes). More efficient but harder to manage.

4. Define delivery in Network Layer.

Delivery means sending packets from source to destination. There are two types:

• Direct delivery: When sender and receiver are on same network (like same office LAN)

• Indirect delivery: When packets must travel through routers to reach different networks (like
sending email to someone in another country)

5. Define Token Bucket.

Token Bucket is like a traffic control system for data:

• Imagine a bucket that fills with "tokens" at a fixed rate

• To send data, you need tokens (1 token per byte, for example)

• If bucket is empty, you must wait for more tokens

• This prevents network flooding while allowing bursts of data when tokens are available
www.gradamic.com

6. What are headers and trailers? How are they added and removed?

• Headers: Extra information added at the beginning of data (like address labels on a package)

• Trailers: Extra information added at the end (like a checksum to detect errors)

• Adding/Removing:

o When sending: Each layer adds its own header (like wrapping a gift in multiple boxes)

o When receiving: Each layer removes its header (like unwrapping the boxes one by
one)

o Example: Transport layer adds port numbers, network layer adds IP addresses

7. What is count-to-infinity problem?

This is a routing issue where routers keep increasing hop counts forever:

• Imagine Router A thinks path to Router D is through Router B (cost=2)

• If link to B breaks, B might say "I can reach D through A (cost=3)"

• Then A updates to cost=4, B updates to cost=5, and so on forever

• Solution: Set a maximum limit (like 16 hops) or use better algorithms

8. What is time-to-live or packet lifetime?

TTL is like an expiration date for packets:

• Every packet starts with a TTL number (usually 64 or 128)

• Each router decreases it by 1 when forwarding

• If TTL reaches 0, packet is discarded (prevents infinite looping)

• Like a "hot potato" game where the packet gets too "cold" to keep passing

9. What is the difference between network layer and transport layer delivery?

• Network layer: Focuses on getting packets from one computer to another using IP addresses
(like postal service delivering to a house)

• Transport layer: Focuses on getting data to the right program on that computer using port
numbers (like delivering to the right person in that house)

10. Define logical addressing.

Logical addressing uses artificial numbers to identify devices:

• IP addresses (like 192.168.1.1) are logical - they can change

• Different from physical (MAC) addresses burned into hardware

• Allows flexible network organization (you can change IPs without changing hardware)
www.gradamic.com

11. Explain the role of registered ports.

Ports are like doors to services on a computer:

• Well-known ports (0-1023): Standard services (HTTP=80, HTTPS=443)

• Registered ports (1024-49151): Assigned to specific apps (MySQL=3306)

• Dynamic ports (49152-65535): Temporary use for private connections

Registered ports help avoid conflicts when installing new services.

12. Define datagrams in switching.

Datagrams are independent packets in connectionless networks:

• Each packet carries full address information

• Packets may take different routes and arrive out of order

• Like sending many letters separately instead of one big package

• Used in IP networks (UDP uses datagrams)

13. Write IP ranges of Class A, B, C, and D.

• Class A: 1.0.0.1 to 126.255.255.254 (for huge networks)

• Class B: 128.1.0.1 to 191.255.255.254 (medium networks)

• Class C: 192.0.1.1 to 223.255.254.254 (small networks)

• Class D: 224.0.0.0 to 239.255.255.255 (multicast groups)

14. List three main functionalities of transport layer.

1. Segmentation: Breaking data into smaller pieces

2. Connection control: Establishing/managing connections (TCP)

3. Error control: Checking for and fixing errors in transmission

15. Define QoS (Quality of Service).

QoS is like VIP treatment for important network traffic:

• Prioritizes certain data (like video calls over file downloads)

• Manages bandwidth, delay, jitter, and packet loss

• Example: Giving Zoom priority over Netflix so calls stay clear

16. Explain HTTP. State the difference between HTTP and HTTPS.

• HTTP: Basic web page transfer protocol (sends data in plain text)

• HTTPS: Secure version that encrypts data using SSL/TLS

• Difference: HTTPS is like sending a sealed letter instead of a postcard

www.gradamic.com

17. Define RST flag in TCP header.

RST (Reset) flag is an emergency "stop" button for TCP:

• Immediately terminates a connection when there's a problem

• Like hanging up the phone suddenly when the call gets messed up

• Used when ports are closed or connections need to abort

18. Define connection-less and connection-oriented services.

• Connection-oriented (TCP): Like a phone call - setup first, then talk, then proper hangup

• Connection-less (UDP): Like mailing letters - just send and hope they arrive

19. Write a note on round-trip time (RTT).

RTT is the time for a signal to go to destination and back:

• Important for determining network speed and timeout values

• Measured using ping commands

• Affected by distance, congestion, and medium (fiber vs satellite)

20. Define bit rate and baud rate.

• Bit rate: Actual number of bits transmitted per second (bps)

• Baud rate: Number of signal changes per second

• They're equal for simple signals, but differ with complex encoding

21. What is ICMP? Mention its applications.

ICMP is the network's "error messaging system":

• Used by ping and traceroute tools

• Reports errors like "destination unreachable"

• Handles network diagnostics and problem reporting

22. Describe piggybacking.

Piggybacking combines two messages into one:

• When computer A sends data to B, it can include ACK for B's previous message

• Saves bandwidth by avoiding separate ACK packets

• Like writing "Got your letter" when sending your own letter
www.gradamic.com

23. What is Telnet?

Telnet is an old remote login protocol:

• Lets you control another computer over network

• Insecure (sends passwords in plain text)

• Mostly replaced by SSH (secure version)

24. What are the advantages and disadvantages of mesh topology?

Advantages:

• Very reliable (many alternate paths)

• Good for heavy traffic

• No central point of failure

Disadvantages:

• Expensive (lots of cables)

• Hard to setup and manage

• Wastes some connections

25. What is the use of bridges?

Bridges connect two similar networks:

• Filter traffic to reduce unnecessary transfers

• Learn which devices are on which side

• Work at data link layer (using MAC addresses)

26. Describe the count-to-infinity problem with example.

(Already answered in Q7, but more detailed example)

Imagine three routers A-B-C:

1. Normally, A thinks path to C is A-B-C (cost=2)

2. If B-C link breaks, B tells A "I can reach C through you"

3. A updates to A-B-A-C (cost=3)

4. B updates to B-A-B-C (cost=4)

And so on until infinity unless stopped
www.gradamic.com

27. Show error detection in 7-bit Hamming code.

Hamming code adds check bits to detect/correct errors:

1. Number bits (b1-b7) where positions 1,2,4 are check bits

2. Each check bit covers certain positions:

o b1: 1,3,5,7

o b2: 2,3,6,7

o b4: 4,5,6,7

3. Receiver checks parity to find which bit is wrong

28. Compare OSI and TCP/IP.

Feature OSI Model TCP/IP Model

Layers 7 layers 4 layers

Development Theoretical Practical

Protocols Generic Specific (TCP/IP)

Adoption Less used Widely used

Session/Presentation Separate Combined with Application

29. Explain transmission delay in flow control.

Transmission delay is the time to push all bits into the wire:

• Formula: (Packet size) / (Bandwidth)

• Example: 1MB file on 1Mbps link takes ~8 seconds

• Flow control manages this to prevent sender overwhelming receiver

www.gradamic.com

SECTION B: Medium-Length Conceptual/Calculative

1. Define TCP/IP Model in detail.

The TCP/IP model has 4 layers:

1. Network Access Layer:

o Combines physical and data link functions

o Handles hardware addressing (MAC)

o Examples: Ethernet, Wi-Fi

2. Internet Layer:

o Uses IP for logical addressing

o Routes packets between networks

o Protocols: IP, ICMP, ARP

3. Transport Layer:

o Manages end-to-end connections

o Protocols: TCP (reliable), UDP (fast)

o Handles flow control and error checking

4. Application Layer:

o User-facing services

o Protocols: HTTP, FTP, DNS, SMTP

o Combines OSI's application, presentation, session

www.gradamic.com

2. Define IPv4 header format in detail.

IPv4 header has these key fields (20-60 bytes):

1. Version: 4 (for IPv4)

2. Header Length: Size of header

3. Type of Service: QoS priorities

4. Total Length: Packet size

5. Identification: Packet ID for fragmentation

6. Flags: Don't fragment/More fragments

7. Fragment Offset: Position if packet is split

8. TTL: Time to live (hop count)

9. Protocol: TCP(6)/UDP(17) etc.

10. Header Checksum: Error check

11. Source/Destination IP: 32-bit addresses

12. Options: Extra features (rarely used)

3. Define TCP features and TCP header in detail.

TCP Features:

• Connection-oriented

• Reliable delivery

• Flow control

• Error checking

• Ordered data

TCP Header Fields:

1. Source/Destination Port (16 bits each)

2. Sequence/Acknowledgement Numbers (32 bits)

3. Header Length (4 bits)

4. Control Flags (URG, ACK, PSH, RST, SYN, FIN)

5. Window Size (flow control)

6. Checksum (error detection)

7. Urgent Pointer (if URG flag set)

8. Options (variable length)

www.gradamic.com

4. Explain SNMP protocol in detail.

Simple Network Management Protocol manages network devices:

Components:

1. Managed Devices (routers, switches)

2. Agents (software on devices)

3. NMS (Network Management Station)

Operations:

• GET: Request information

• SET: Change configurations

• TRAP: Unsolicited alerts from devices

MIB: Database of manageable objects

Used for monitoring performance, detecting faults, and configuring devices remotely.

5. Define topology. Explain bus, star, and ring topologies with pros and cons.

Topology is how devices are interconnected.

Bus Topology:

• Single cable with all devices attached

• Pros: Simple, cheap

• Cons: Single point of failure, hard to troubleshoot

Star Topology:

• All devices connect to central hub/switch

• Pros: Easy to manage, fault tolerant

• Cons: Hub is single point of failure

Ring Topology:

• Devices form a circle, data passes around

• Pros: Equal access, orderly

• Cons: Break in ring kills network

www.gradamic.com

6. List and explain transmission mediums.

1. Twisted Pair:

o Phone/network cables

o Cheap, flexible

o Limited distance/speed

2. Coaxial Cable:

o Thick center wire

o Better shielding than twisted pair

o Used for cable TV

3. Fiber Optic:

o Glass fibers carrying light

o Very fast, long distance

o Expensive, fragile

4. Wireless:

o Radio waves (Wi-Fi, cellular)

o Convenient

o Security/signal issues

7. Explain encoding types in the physical layer of OSI model.

Encoding converts bits to signals:

1. Unipolar: All signals on one side of zero

2. NRZ (Non-Return to Zero): High=1, Low=0

3. Manchester: Transition in middle of bit

o Down-up=1, Up-down=0

4. Differential Manchester: Transition at start indicates 0

5. AMI (Alternate Mark Inversion): Alternate 1s as +/-

www.gradamic.com

8. Explain noisy channel protocols with diagrams.

Noisy channels need error handling:

1. Stop-and-Wait ARQ:

o Send frame, wait for ACK

o Timeout and resend if no ACK

o Simple but slow

2. Go-Back-N ARQ:

o Window of multiple frames sent

o If error, resend all from lost frame

o More efficient but wastes some

3. Selective Repeat ARQ:

o Only resend bad frames

o Most efficient but complex

9. Write about commands: ipconfig, netstat, ping, hostname, tracert.

• ipconfig: Shows IP configuration (IP, subnet, gateway)

• netstat: Displays network connections/routing tables

• ping: Tests connectivity to another host

• hostname: Shows computer's network name

• tracert: Maps route to destination showing all hops

www.gradamic.com

10. Explain working of TCP and differentiate TCP and UDP with frame format.

TCP Working:

1. 3-way handshake (SYN, SYN-ACK, ACK)

2. Data transfer with sequencing

3. Acknowledgments and retransmissions

4. Flow control using window size

5. Graceful close (FIN)

TCP vs UDP:

Feature TCP UDP

Connection Connection-oriented Connectionless

Reliability Reliable Unreliable

Ordering Ordered Unordered

Speed Slower Faster

Header Size Larger (20+ bytes) Smaller (8 bytes)

Use Cases Web, email Video, DNS

11. Explain ICMP BGP protocol and application.

(Note: ICMP and BGP are separate)

ICMP: Covered earlier (network diagnostics)

BGP (Border Gateway Protocol):

• The "postal service" of the internet

• Routes traffic between autonomous systems (ISPs)

• Uses path-vector algorithm

• Makes routing decisions based on policies

• Critical for internet backbone routing

www.gradamic.com

12. Define ARQ error control.

Automatic Repeat reQuest handles errors:

1. Stop-and-Wait: Send one frame, wait for ACK

2. Go-Back-N: Send window of frames, if error go back N

3. Selective Repeat: Only resend damaged frames

Uses acknowledgments, timeouts, and sequence numbers

13. Explain CSMA/CD with CSMA/CA with diagrams.

CSMA/CD (Ethernet):

1. Listen before transmit

2. If collision detected, stop and wait random time

3. Used in wired networks where collisions can be detected

CSMA/CA (Wi-Fi):

1. Listen before transmit

2. Send small RTS (Request to Send) first

3. Wait for CTS (Clear to Send)

4. Used in wireless where collisions can't always be detected

14. Explain congestion control with labeled diagram.

Congestion control prevents network overload:

1. Slow Start: Exponential increase until threshold

2. Congestion Avoidance: Additive increase

3. Fast Retransmit: On duplicate ACKs

4. Fast Recovery: After retransmit

Diagram would show TCP window size growing then reducing when packet loss occurs.
www.gradamic.com

15. Define and explain Go-Back-N and Selective Repeat protocols.

Go-Back-N:

• Sender can have N unacknowledged packets

• Receiver only accepts in-order packets

• If packet N is lost, all N onward are resent

• Like rewinding and replaying a tape

Selective Repeat:

• Receiver buffers out-of-order packets

• Only lost packets are resent

• More efficient but needs more memory

• Like replacing only damaged books in a set

16. Discuss asymmetric cryptography. Explain RSA algorithm with example (character "F" or custom
input).

Asymmetric crypto uses public/private key pairs:

• Public key encrypts, private key decrypts

• Secure because deriving private key from public is extremely hard

RSA Example (for 'F'=70):

1. Choose primes p=5, q=11

2. n = p*q = 55

3. φ(n) = (5-1)*(11-1) = 40

4. Choose e=3 (coprime with 40)

5. Find d where (d*e) mod φ(n) = 1 → d=27

6. Public key: (3,55), Private key: (27,55)

7. Encrypt: 70³ mod 55 = 343000 mod 55 = 35

8. Decrypt: 35²⁷ mod 55 = 70 (original)

www.gradamic.com

17. Explain Quality-of-Service parameters.

QoS measures:

1. Bandwidth: Capacity of link

2. Delay: Time for packet to travel

3. Jitter: Variation in delay

4. Packet Loss: Percentage lost

5. Throughput: Actual data rate

18. List and explain four network devices and transmission impairment.

Devices:

1. Hub: Dumb repeater (layer 1)

2. Switch: Smart hub using MAC addresses (layer 2)

3. Router: Routes between networks (layer 3)

4. Gateway: Connects different protocol networks

Impairments:

1. Attenuation: Signal weakening over distance

2. Distortion: Signal shape changes

3. Noise: Random interference

4. Crosstalk: Signal leakage between cables

19. List and explain disadvantages of Stop-and-Wait protocol.

1. Inefficient: Only one packet at a time

2. Slow: Long delays for ACKs

3. Wasteful: Bandwidth unused while waiting

4. Poor for long distances: High latency hurts performance

5. No pipelining: Can't overlap transmissions

20. Discuss role of ICMP in network layer.

ICMP is the network's "error reporting system":

1. Reports delivery problems ("Destination Unreachable")

2. Provides diagnostic tools (ping, traceroute)

3. Handles flow control ("Source Quench")

4. Supports router discovery

5. Manages time exceeded messages (TTL=0)

www.gradamic.com

SECTION C: Long Answer / Analytical / Problem-Solving

1. Describe all layers of OSI model with labelled diagram.

OSI Model Layers:

1. Physical Layer:

o Raw bit transmission

o Cables, connectors, voltages

o Units: Bits

2. Data Link Layer:

o Error detection/correction

o MAC addressing

o Units: Frames

3. Network Layer:

o Logical addressing (IP)

o Routing

o Units: Packets

4. Transport Layer:

o End-to-end connections

o Flow control

o Units: Segments

5. Session Layer:

o Manages dialogues

o Synchronization

o Units: Data

6. Presentation Layer:

o Data translation

o Encryption/compression

o Units: Data

7. Application Layer:

o User interfaces

o Network services
www.gradamic.com

o Units: Data

Diagram would show vertical stack with encapsulation process.

2. Differentiate between network topologies with diagrams.

(Expanding on Section B Q5 with more topologies)

Mesh:

• Every device connects to every other

• Most reliable but most expensive

Tree:

• Hierarchical star networks

• Good for large organizations

• Single root failure breaks network

Hybrid:

• Combination of other topologies

• Balances cost and reliability

Diagrams would show physical layouts of each type.

3. Explain various network devices layer-wise on OSI model.

Layer 1 (Physical):

• Hub: Repeats signals to all ports

• Repeater: Boosts signals

Layer 2 (Data Link):

• Switch: Forwards based on MAC

• Bridge: Connects two LANs

Layer 3 (Network):

• Router: Routes between networks

• Layer 3 switch: Switch with routing

Layer 4+:

• Gateway: Translates between protocols

• Firewall: Filters traffic

• Proxy: Intermediary for requests

www.gradamic.com

4. Define and compare LAN standards.

Ethernet (IEEE 802.3):

• Wired LAN

• CSMA/CD

• Speeds: 10Mbps to 100Gbps

Wi-Fi (IEEE 802.11):

• Wireless LAN

• CSMA/CA

• Standards: 802.11a/b/g/n/ac/ax

Token Ring (IEEE 802.5):

• Older LAN tech

• Token passing

• Mostly obsolete

Comparison focuses on speed, cost, and use cases.

5. Explain random access protocols.

Methods for shared medium access:

1. ALOHA:

o Pure ALOHA: Send anytime, retry if collision

o Slotted ALOHA: Send at time slot starts

2. CSMA:

o Listen before transmit

o Variants: 1-persistent, p-persistent

3. CSMA/CD:

o Ethernet's method

o Detect collisions while transmitting

4. CSMA/CA:

o Wi-Fi's method

o Avoid collisions with RTS/CTS

www.gradamic.com

6. Describe interdomain and intradomain routing algorithms.

Intradomain (within AS):

• RIP: Distance vector, hop count metric

• OSPF: Link state, Dijkstra's algorithm

• EIGRP: Cisco's advanced distance vector

Interdomain (between ASes):

• BGP: Path vector, policy-based

• Makes routing decisions based on paths and policies

7. Explain three-way handshaking for connection management.

TCP connection setup:

1. SYN: Client sends SYN packet with initial sequence number

2. SYN-ACK: Server responds with SYN and ACK (client's SYN+1)

3. ACK: Client acknowledges server's SYN

This synchronizes sequence numbers and establishes the connection.

8. Describe cryptography and RSA algorithm with an example.

(Expanding on Section B Q16)

Cryptography Types:

1. Symmetric: Same key (AES, DES)

2. Asymmetric: Key pairs (RSA, ECC)

3. Hash: One-way (SHA, MD5)

RSA Steps:

1. Key generation (as before)

2. Encryption: c = m^e mod n

3. Decryption: m = c^d mod n

4. Security relies on factoring n being hard

www.gradamic.com

9. Explain IPv4 vs IPv6 differences.

Feature IPv4 IPv6

Address Size 32 bits (4B) 128 bits (16B)

Notation Dotted decimal Hexadecimal

Header Complex (20-60B) Simpler (40B fixed)

Fragmentation Sender/router Sender only

Security Optional (IPsec) Built-in IPsec

Configuration Manual/DHCP Auto-config

Example 192.168.1.1 2001:0db8:85a3::8a2e:0370:7334

10. Solve: CRC using polynomial (e.g., x³ + x + 1) for data like 10011101, 110101, etc.

CRC steps for data 110101 with poly x³+x+1 (1011):

1. Append 3 zeros: 110101000

2. Divide by 1011 using XOR:

Copy

Download

110101000

1011

----- XOR

1100

1011

---- XOR

1110

1011

---- XOR

1010
www.gradamic.com

1011

---- XOR

001 (remainder)

3. Append remainder: 110101001

11. Explain error control in data link layer.

Error handling methods:

1. Error Detection:

o Parity bits

o Checksums

o CRC

2. Error Correction:

o Hamming codes

o Reed-Solomon

3. Retransmission:

o ARQ protocols

o Sequence numbers

o Acknowledgments

12. Divide given IP network (e.g., 200.1.2.0) into subnets.

Example: 200.1.2.0/24 into 4 subnets:

1. New mask: /26 (2 extra bits)

2. Subnet size: 64 addresses (62 usable)

3. Subnets:

o 200.1.2.0/26

o 200.1.2.64/26

o 200.1.2.128/26

o 200.1.2.192/26
www.gradamic.com

13. Describe application and session layer functions in OSI model.

Session Layer:

• Establishes/manages/terminates connections

• Synchronization (checkpoints)

• Dialog control (half/full duplex)

Application Layer:

• User interface to network

• Provides services (email, file transfer)

• Protocols: HTTP, FTP, SMTP, DNS

14. Explain DNS, FTP, SMTP, DNS, ARP, HTTP protocols.

DNS: Converts names to IPs (phonebook)

FTP: File transfers (port 20/21)
SMTP: Email sending (port 25)
ARP: Finds MAC from IP
HTTP: Web page transfer (port 80)

15. Calculate total time for transmission (given data size, bandwidth, headers).

Example: 1MB file, 1Mbps link, 40B header:

1. Total data = 1MB + headers

2. Transmission time = total bits / bandwidth

3. Propagation time = distance/speed

4. Total time = transmission + propagation + processing

16. Sliding window: calculate window size and sequence number (given Tp, Tt).

Window size ≥ 1 + 2*(Tp/Tt) for efficiency

Sequence numbers must cover window (e.g., 3-bit for window 7)
www.gradamic.com

17. Compare TCP and UDP in context of headers.

(Expanding on Section B Q10)

TCP header (20-60B) vs UDP header (8B fixed)
TCP has sequence numbers, ACKs, window size, etc.

18. Calculate efficiency (ALOHA, Stop-and-Wait, etc.) or fault-based degradation.

Pure ALOHA: 18% max efficiency

Slotted ALOHA: 36% max
Stop-and-Wait: 1/(1+2a) where a = Tp/Tt

19. Differentiate between BOOTP and DHCP.

BOOTP:

• Older protocol

• Static IP assignments

• No lease time

DHCP:

• Dynamic IP assignments

• Lease management

• Additional options

20. Encode bitstreams using line coding schemes (Unipolar, NRZ-L, Manchester, etc.)

Example: Encoding 1011001

Unipolar: High=1, Low=0 (DC component)

NRZ-L: Level changes at bit changes
Manchester: Transition in middle (IEEE)
Differential Manchester: Transition at start for 0