Unit II – SENSORS, ACTUATORS AND

WIRELESS SENSOR NETWORK (WSN)

LU 18 – WIRELESS SENSOR NETWORKS

TECHNOLOGY
 Introduction
• A set of sensors can be networked using a wireless system.
• They cooperatively monitor the physical and environmental conditions,
such as temperature, sound, vibration, pressure, motion, or hazardous
gas-leaks and pollutants at different locations.
• A WSN acquires data from multiple and remote locations. For example,
in Internet of Waste Containers, the sensors wirelessly communicate the
waste containers statuses in a waste management system in a smart
city; sensors in Internet of Streetlights acquire data of ambient light
conditions; and a WSN communicates the nearby traffic densities data
for the control and monitoring of traffic signals.
• Each node of the WSN has an RF transceiver. The transceiver functions
as both, a transmitter and receiver.
• WSN infrastructure, data-link layer MAC protocols, routing protocols,
various integration approaches, challenges of security, QoS and
configuration, and WSN specific IoT applications.
 WSN Concepts
• WSN is defined as a network in which each sensor node connects
wirelessly and has the capability of computation, for data compaction,
aggregation and analysis.
• Each one also has communication as well as networking capabilities. A
WSN consists of spatially distributed autonomous devices.
• WSN History
• Distributed Sensor Networks (DSN) using network communication started.
• Many spatially distributed sensing nodes collaborate and network
autonomously on best effort basis.
• Research and applications of Low Power Wireless Integrated Micro-
Sensors (LWIM) use in DSNs.
• WSNs have a large number of IoT applications. Examples are smart homes
and smart city.
• The WSN nodes can sense and communicate, using the Internet, wirelessly
the data from remote locations, such as industrial plant machines, forests,
lakes, gas or oil pipelines, which may sometimes not be easily accessible.
 Context-based Node Operations
• The dictionary meaning of the word ‘context’ is ‘the circumstances that
form the setting of an event, statement, or idea, and in terms of which it
can be fully understood.
• A WSN node can adapt, re-program or do another task using a sensor
and associated circuit, computations, networking capabilities and
context at that node.
• When nodes can adapt their operations in response to changes in the
environment, then it is called context-dependent sensing, networking
and computing.
• The application layer programs help the node to differentiate the tasks
to be undertaken on a changed context.
 Context-based Node Operations
• Context can be a physical, computing, user, structural or temporal
context. Following may correspond to the context for re-programming
of the actions of the WSN nodes:
– Past and present surrounding situations
– Actions such as the present network
– Surrounding devices or systems
– Changes in the state of the connecting network
– Physical parameters such as present time of the day
– Nearest connectivity currently available
– Past sequence of actions of the device user
– Past sequence of application or applications
– Previously cached data records
– Remaining memory and battery power at present
 WSN Architecture
• The three layers are application layer, network layer (serial link with
data-link MAC), physical cum data-link layer (MAC + Physical Layer).
 WSN Architecture
• The application layer software components are sensor management,
sensor query and data dissemination, task assignment, data
advertisement and application-specific protocols.
• Sensor, CPU and program sensor node constitute the application and
network layers.
• Network layer links serially to the data-link layer, and may include the
coordination or routing software.
• A serial link interconnects the layers to a wireless radio circuit and
antenna.
• The radio circuit is at physical cum data-link layer. Communication
subsystem uses MAC and physical protocols.
 Architecture for Connecting Nodes
• Two architectures for connecting WSN nodes, fixed connecting infrastructure
of WSN nodes, coordinators, relays, gateways and routers, and mobile adhoc
network of WSNs, access points, routers, gateways and multi-point relays.
 Architecture for Connecting Nodes – cntd.,
• An access point is a fixed point transceiver to provide the accessibility
to nodes present nearby or nodes reaching in the wireless range.
• A multipoint relay connects to other networks such as the Internet or
mobile service provider network.
• A router’s role is to select a path for packet transmission among the
presently available paths in the network.
• A coordinator provides the link between the two networks.
• Fixed infrastructure example is a smart home network consisting of
WSNs at security surveillance points, refrigerator, air conditioner,
microwave, TV and computers with Wi-Fi access point.
• Another example is WSN nodes, access points, routers, gateways and
multi-point relays. At an industrial plant, multiple machine locations,
stores, offices, sales and receipts and other plant locations.
Wireless Multi-Hop Infrastructure Network Architecture

• MINA is a layered architecture. The WSN nodes have data sensing as

well as capabilities of forwarding towards the access point (base
station).
• The nodes can be mobile and have coverage and mobility range for
communication to remote access points.
• Access points possess data gathering and processing capabilities and
have connectivity with a larger network, such as the Internet.
• Each node connects to a short-distant neighbour. When the node moves
to longer distances then it communicates through 2 or 3 hops to the
access point (base station).
• Each node has low-power transceivers to the nearest neighboring layer
WSNs.
Layered Architecture for network of nodes
 Layered Architecture for network of nodes
• Assume that the base station is surrounded by three layers of WSNs.
• Layer 1 WSNs directly connect. Layer 2 WSNs first connect to layer 1
WSNs functioning as coordinators and then directly connect.
• Layer 3 WSNs first connect to layer 2 WSNs functioning as
coordinators, then connect to layer 1 WSNs and then connect to the
access point.
• The figure shows that layer 1 WSN1 and WSN6 connect directly to the
access point.
• It means hop count = 1. The figure also shows WSN 2 and WSN 3 at
layer 2. They connect by one hop to WSN1 and next hop through
WSN1. Hop count = 2 for layer 2. It means that WSN 2 connects to
WSN1 which has connectivity with the access point (base station).
 Layered Architecture for network of nodes
• The figure also shows WSN 4 and WSN 5 at layer 3. They connect by
three hops, one to layer 2 WSN and then layer 1 WSN and then to the
access point.
• Hop count = 3 for layer 3. It means that WSN 5 connects to WSN2,
then to WSN1 and then the access point (base station).
• WSN 4 connects to WSN3, then to WSN1 and then the access point
(base station).
• Access points connect the clusters using wireless LAN (802.11b)
protocol.
• The access points enable connectivity to the Internet.
 Multiple Clusters Architecture
• Each cluster has a gateway node. A set of clusters with a gateway each
has one cluster with a cluster-head gateway.
• Multi-cluster architecture has a number of clusters, which associate a
cluster-head gateway.
• Cluster-head enables a tree-like topology of the clusters in a multi-
cluster architecture.
• The cluster formation and election of cluster-heads is autonomous in
distributed WSNs and WSN clusters.
• The figure shows two clusters and also shows a cluster-head gateway.
• Head gateway connects to the Internet or cellular of other networks and
provides connectivity to WSNs in the multiple clusters.
Multi-cluster architecture for network of nodes
 Multiple Clusters Architecture
• A number of clusters associated with the cluster-head gateway depends
on the coverage required in the network.
• When a cluster-head exists then a gateway interconnects two clusters.
Each node connects to a short-distant neighbor.
• The node connects to a gateway node through one, two or more hops
when it connects to the WSN of another cluster.
• When a node moves to longer distances then it communicates to the
neighboring cluster through the gateway.
• Cluster 1 is an ad-hoc network of mobile WSNs (WSN 1 to WSN5). The
WSNs in a cluster may have mesh architecture or layered architecture.
Each WSN connects to another WSN in its wireless range. Cluster 2 is
an ad-hoc network of mobile WSNs (WSN 6 to WSN10).
• Clustered architecture enables data compaction or fusion and
aggregation. A gateway communicates compacted or aggregated data to
another cluster. A cluster-head further aggregates, compacts or fuses the
data before communication to the web server or cloud using the Internet.
 WSN Protocols
• Network protocols have design goals as:
• (i) limit computational needs,
• (ii) limit use of battery power and thus bandwidth limits, operation in
self-configured ad-hoc setup mode
• (iii) limit memory requirement of the protocol.
Data-link Layer Media Access Control (MAC) Protocol:
• S-MAC (Sensor-MAC) protocol can be deployed at the data-link layer.
• S-MAC nodes go to sleep for prolong periods. They need to synchronize
at periodic intervals.
• S-MAC protocol enables use of the energy-efficient, collision-free
transmission, and intermittently synchronizes the operations.
• Collision-free transmission results from scheduling of a channel. Each
node can be allocated the channel.
• Channels are reused such that collision-free transmission takes place and
retransmission is not resorted to.
 WSN Protocols
Routing Protocols:
• The network layer uses multi-hop route determination, energy-efficient
routing, route caching and directed diffusion of data.
• Routing protocols are either proactive or reactive. Proactive protocols
keep route cache and determine the route in advance.
• Reactive protocols determine the route on demand.
• Routing protocols are table-driven when a routing table guides the paths
available.
• Routing protocols are demand-driven when a source demands the route
and it guides the paths available.
• Fixed Wireless Sensor Networks may use Cluster-head Gateway Switch
Routing (CGSR) which guides the paths using heuristic routing schemes.
 WSN Infrastructure Establishment
• Sensors with associated CMOS low power ASIC circuits, their radio ranges and
energy-efficient coding.
• Infrastructure selection, (i) fixed connecting infrastructure of nodes,
coordinators, relays, gateways and routers, or (ii) mobile ad-hoc network of
mobile WSNs with limited or unspecified mobility region.
• Network topology and architecture according to the applications and services,
wireless multi-hop infrastructure network architecture or multi-cluster
architecture.
• Network nodes self-discovering, self-configuring and self-healing protocols,
localization, mobility range, security, data-link and routing protocols, link
quality indicator (LQI) (packet reception ratio), coverage range, and QoS
required according to applications and services.
• Clusters, cluster gateways, cluster-heads and clusters hierarchy.
• Routing, data aggregation, compaction, fusion and direct diffusion.
• Synchronization of times at intermittent intervals as time is used as reference in
internode distance estimation, accounting for the delays, localization, ranging
and location services.
 WSN Various Integration Approaches
WSN needs integrated approaches for:
• Nodes design and provisioning of resources
• Nodes localization
• Nodes mobility
• Sensor connecting architecture
• Sensor networking architectures
• Data dissemination protocols
• Security protocols
• Data-link layer and routing protocols
• Integration with sensors data, other than wireless sensors data, for IoT
applications and services
 Quality of Service
• Quality of Service (QoS) is an average weighted QoS metric over the
lifetime of a network.
• Metrics:
(i) Average delay: Measure of the time taken in generation of sensor data
and its delivery up to the destination
(ii) Lifetime: Time up to which the WSN functions effectively or given
energy resources of the modes will last
(iii) Throughput: Bytes per second delivered up to the destination. Low
throughput means high delays. It also relates to the bandwidth of the
network
(iv) Link Quality Indicator (LQI) which means packets delivered/packet
transmitted from the nodes.
 Quality of Service
Challenges:
• routing through nodes of high throughputs, lower delay and paths which
use low energy resources
• Link
• maintaining the product of priority and delay, which means higher
priority packets take lower delay paths and lower priority packets take
higher delay paths.
• A metric is the coverage of a sensor network. The coverage depends on
the density and locations of the nodes in the region, communication
range and sensitivity of the nodes.
• Another metric is percentage of times the network cover occurrence of
an event in case of surveillance systems, hazard chemicals or fire-
detection systems.
 Configuration
• Challenges of configuring in view of resource constraints with the nodes
statically or dynamically or self-automatic are the following:
• (i) locations and mobility range of the WSN nodes,
• (ii) clusters,
• (iii) gateways,
• (iv) cluster-heads,
• (v) sampling rate of the sensed parameters
• (vi) their aggregation, compaction and fusion.
• Challenges are to be met with limited battery power, storage,
computation capability, bandwidth and scalability limit on the nodes.
 WSN Nodes Secure Communication
• Sensor networks need secure communication for data privacy and
integrity.
• Authentication ensures data from the sensing node only, maintains data
integrity and disables the communication of messages from
unauthenticated sources.
• Privacy ensures data secrecy and eavesdropping.
• SPINS (Security Protocols in Network of Sensors) use the symmetric
cryptographic protocol because of the following reasons—an
asymmetric cryptographic method which has high overhead in term of
memory and computations for digital signatures, key-generation and
verification.
• SPINS is a suite of security protocols for the sensor networks which are:
(i) Secure Network Encryption Protocol (SNEP)
• (ii) micro-Tesla (m-Tesla). SPINS use block cipher encryption functions.
 WSN Nodes Secure Communication
• SNEP enables secure point-to-point communication. It ensures data privacy
and integrity. It secures communication after authentication process. It does
not require message replay, thus message remains fresh.
• Access point distributes session key to A and B using SNEP. Two nodes, A and
B, share six keys.
• The encryption keys, KAB and KBA, Cipher-block-chaining Message
Authentication Code (MAC) keys, KAB and KBA, and counter keys, CA, CB are
the shared keys. MAC ensures message integrity and privacy.
• m-TESLA is light-weight version of TESLA. It enables authenticated
broadcasting. The authentication is micro-timed and efficient. It results in
stream-loss tolerant secure authentication.
• An access point authenticates using m-TESLA. First a packet is listened and
considered as a parent. It is authenticated later. This makes secure
authentication stream-loss tolerant.
• Distributed sensor networks use a key management scheme, which can be (i)
based on probabilistic key sharing or (ii) random key pre-distribution key
sharing.
 WSN Nodes Secure Communication
Localized Encryption and Authentication Protocol (LEAP)
• Different packets when using LEAP use different keying mechanisms.
Security needs decide the mechanism. A node uses four keys: individual
key, group key, cluster key and share a pair of key with the neighbor.
• A challenge for highly-distributed architecture with localized
coordination is the implementation of goals of a system for applications
and services.
• Implementation needs are autonomous operation, self-organisation, self-
configuration, adaptation, energy conservation at physical, MAC, link,
route, application layer, design of scalable node density, number and
types of networks.
• Network is data-centric network and route nodes have no addressability.
 WSN Nodes Secure Communication
Localized Encryption and Authentication Protocol (LEAP)
• Three challenges, viz. security, QoS and configuring of nodes are as
follows:
Security Challenges are:
• Hello flood attack: An attacker node sends hello messages repeatedly, and
thus drains the energy of the attacked node.
• Sybil attack is an attack where a single node, presents itself as different
entities at different times.
• Selective forwarding attack is an attack when the attacker node does not
forward the attacked node messages on receiving.
• Sinkhole attack is when an attacked node behaves as an access point and
receives the messages without forwarding them.
• Wormhole attack is an attack where the attacker node gives false
information of distances of the destinations, thus forcing the attacked node
to take longer paths. The longer path has high latency and thus high
delivery delays of packets.
 WSN IoT Applications
• Mobile, tablets, IP-TV, VOIP telephony, video conferencing, video-on-
demand, video-Surveillance, Wi-Fi and Internet
• WSN nodes and wireless actuator nodes, which can be built using
ZigBeeIP are nodes for home security access control and security alerts,
lighting control, home health care, fire detection, leak detection, energy
efficiency, solar panel monitoring and control, temperature monitoring and
HVAC control and automated meter reading.
• Figure shows source-end cluster of ZigBee WSN and actuator nodes,
coordinators and routers connected through gateway and set of RPL routers
for data packets from IPv6 addresses and communicating with IoT and
M2M IoT applications and services layer for data packets from IPv6
addresses.
• ZigBee devices form WPAN (Wireless Personal Area Network) devices in
home network. WSN nodes, coordinating nodes and routing nodes can also
use ZigBee protocol for physical layer wireless connectivity. ZigBee is
IEEE 802.15.4 standard protocol for physical/data-link layer (Phy/MAC).
 WSN IoT Applications
• ZigBee IP is an enhancement for IPv6 connectivity. ZigBee IP is
enhancement with RFD. RFD (Reduced Function Device) means one
that functions for the ‘sleepy’ battery-operated device.
• Sleepy means one that wakes up infrequently, sends data and then goes
back to sleep. 6lowpan refers to IPv6 over low power wireless personal
area network which permits communication of IPv6 packets in a sensor
network.
WSN IoT Applications