SRI CHANDRASEKHARENDRA SARASWATHI VISWA MAHAVIDYALAYA

(University established under section 3of UGC Act 1956) (Accredited with ‘A’ Grade by NAAC)

Enathur, Kanchipuram – 631 561

DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING

Course Material for

Wireless Sensor Networks

FULL TIME B.E., IV YEAR / VII SEMESTER
Prepared By:Dr.S.Omkumar,ECE
Approved by: Prof.V.Swaminathan
PEC3 WIRELESS SENSOR NETWORKS VII SEMESTER

PRE-REQUISITE:
Basic knowledge of Data Communication Networks
OBJECTIVES:
 To understand the basics of Wireless sensor Networks
 To learn the Architecture of WSN
 To understand the concept of Networking and Networking in WSN

UNIT I OVERVIEW OF WIRELESS SENSOR NETWORKS (9 Hrs)

Single-Node Architecture - Hardware Components- Network Characteristics- unique constraints
and challenges, Enabling Technologies for Wireless Sensor Networks- Types of wireless sensor
networks.
UNIT II ARCHITECTURES (9 Hrs)
Network Architecture- Sensor Networks-Scenarios- Design Principle, Physical Layer and
Transceiver Design Considerations, Optimization Goals and Figures of Merit, Gateway
Concepts, Operating Systems and Execution Environments- Introduction to TinyOS and nesC-
Internet to WSN Communication
UNIT III NETWORKING SENSORS (10 Hrs)
MAC Protocols for Wireless Sensor Networks, Low Duty Cycle Protocols And Wakeup
Concepts - SMAC, - B-MAC Protocol, IEEE 802.15.4 standard and ZigBee, the Mediation
Device Protocol, Wakeup Radio Concepts, Address and Name Management, Assignment of
MAC Addresses, Routing Protocols Energy-Efficient Routing, Geographic Routing.
UNIT IV INFRASTRUCTURE ESTABLISHMENT (8 Hrs)
Topology Control, Clustering, Time Synchronization, Localization and Positioning, Sensor
Tasking and Control.
UNIT V SENSOR NETWORK PLATFORMS AND TOOLS (9 Hrs)
Sensor Node Hardware – Berkeley Motes, Programming Challenges, Node-level software
platforms, Node level Simulators, State-centric programming.

OUTCOMES: Total: 45 Hrs

Upon completion of the course, students will be able to:
 Understand challenges and technologies for wireless networks
 Understand architecture and sensors
 Establishing infrastructure and simulations
TEXT BOOKS:
1. Holger Karl & Andreas Willig, "Protocols and Architectures for Wireless Sensor
Networks”, John Wiley, 2005.
2. Feng Zhao & Leonidas J.Guibas, “Wireless Sensor Networks-An Information Processing
Approach", Elsevier, 2007
3. Waltenegus Dargie , Christian Poellabauer, “Fundamentals Of Wireless Sensor Networks
- Theory And Practice”, By John Wiley & Sons Publications, 2011
REFERENCES:
1. KazemSohraby, Daniel Minoli, & TaiebZnati, “Wireless Sensor Networks-Technology,
Protocols, and Applications”, John Wiley, 2007.
2. Anna Hac, “Wireless Sensor Network Designs”, John Wiley, 2003
Wireless Sensor Networks

Unit I / Overview of WSN

Prepared
By

Dr.S.Omkumar/Associate Prof
Department of ECE, SCSVMV

1
 Syllabus / Unit - I
• Overview of WSN:
• Single-Node Architecture - Hardware
Components - Network Characteristics - Unique
constraints and challenges - Enabling
Technologies for Wireless Sensor Networks-
Types of wireless sensor networks.

2
 Topic 1

Introduction to WSN

3
 Introduction
• A Sensor is a device used to gather information about a
physical process and translate into electrical signals that
can be processed, measured and analyzed.
• The physical process can be any real-world information
like temperature, pressure, light, sound, motion,
position, flow, humidity, radiation etc.
• A Sensor Network is a structure consisting of sensors,
computational units and communication elements for
the purpose of recording, observing and reacting to an
event or a phenomenon.
• The events can like physical world, an industrial
environment, a biological system while the controlling
or observing body can be a consumer application,
government, civil, military, or an industrial entity.
4
• Such Sensor Networks can be used for remote sensing,
medical telemetry, surveillance, monitoring, data
collection etc.

5
 Wireless Sensor Networks
• A typical sensor network consists of sensors, controller
and a communication system. If the communication
system in a Sensor Network is implemented using a
Wireless protocol, then the networks are known as
Wireless Sensor Networks.

6
• According to technologists, Wireless Sensor Networks
is an important technology for the twenty first century.
• Recent developments in MEMS Sensors (Micro Electro
Mechanical System) and Wireless Communication has
enabled cheap, low power, tiny and smart sensors,
deployed in a wide area and interconnected through
wireless links for various civilian and military
applications.
• A Wireless Sensor Network consists of Sensor Nodes
deployed in large quantities and support sensing, data
processing, embedded computing and connectivity.

7
 Motivation for WSN
• The recent developments in engineering,
communication and networking led to new sensor
designs, information technologies and wireless
systems.
• Such advanced sensors can be used as a bridge
between the physical world and the digital world.
• Sensors are used in numerous devices, industries,
machines and help in avoiding infrastructure failures,
accidents, conserving natural resources, preserving
wildlife, increase productivity, provide security etc.
• The use of distributed sensor network contributed by
the technological advances in VLSI, MEMS and Wireless
Communication.
8
• With the help of modern semiconductor technology,
powerful microprocessors can be developed, smaller in
size when compared to the previous generation
products. This miniaturization of processing,
computing and sensing technologies led to tiny, low-
power and cheap sensors, controllers and actuators.

9
 Elements of WSN
• A typical wireless sensor network can be divided into
two elements. They are:
– Sensor Node
– Network Architecture
• A Sensor Node in a WSN consists of four basic
components. They are:
– Power Supply
– Sensor
– Processing Unit
– Communication System

10
Fig 2 / Basic Components of WSN

11
 Elements of WSN (Cont)
• The sensor collects the analog data from the physical
world and an ADC converts this data to digital data.
• The main processing unit a microprocessor or a
microcontroller, performs an intelligent data processing
and manipulation. Communication system consists of
radio system, a short-range radio for data transmission
and reception.
• As all the components are low-power devices, a small
battery like CR-2032, is used to power the entire
system.
• A Sensor Node consists of not only the sensing
component but also other important features like
processing, communication and storage units.
12
• With all these features, components and
enhancements, a Sensor Node is responsible for
physical world data collection, network analysis, data
correlation and fusion of data from other sensor with
its own data.

13
 Network Architecture
• When a large number of sensor nodes are deployed in
a large area to monitor a physical environment, the
networking of these sensor nodes is equally important.
A sensor node in a WSN not only communicates with
other sensor nodes but also with a Base Station (BS)
using wireless communication.

14
• The base station sends commands to the sensor nodes
and the sensor node perform the task by collaborating
with each other.
• The sensor nodes in turn send the data back to the
base station. A base station also acts as a gateway to
other networks through the internet.
• After receiving the data from the sensor nodes, a base
station performs simple data processing and sends the
updated information to the user using internet.
• If each sensor node is connected to the base station, it
is known as Single-hop network architecture.
• Although long distance transmission is possible, the
energy consumption for communication will be
significantly higher than data collection and
computation.
15
Fig 4 / Single Hop Architecture

16
 Multi-hop Architecture
• Hence, Multi-hop network architecture is usually used.
Instead of one single link between the sensor node and
the base station, the data is transmitted through one
or more intermediate node.

17
• This can be implemented in two ways. Flat network
architecture and Hierarchical network architecture.
• In flat architecture, the base station sends commands
to all the sensor nodes but the sensor node with
matching query will respond using its peer nodes via a
multi-hop path.
• In hierarchical architecture, a group of sensor nodes
are formed as a cluster and the sensor nodes transmit
data to corresponding cluster heads.
• The cluster heads can then relay the data to the base
station

18
Fig 6 / Flat and Hierarchical Network
Architectures

19
 Network Topologies in WSN
• A WSN can be either a single-hop network or a multi-
hop network. The following are a few different network
topologies that are used in WSNs.
• Star Topology
• In star topology, there is a single central node known as
hub or switch and every node in the network is
connected to this hub. Star topology is very easy to
implement, design and expand. The data flows through
the hub and plays an important role in the network and
a failure in the hub can result in failure of entire
network.

20
• Tree Topology
• A tree topology is a hierarchical network where there is
a single root node at the top and this node is
connected to many nodes in the next level and
continues. The processing power and energy
consumption is highest at the root node and keeps on
decreasing as we go down the hierarchical order.

• Mesh Topology
• In mesh topology, apart from transmitting its own data,
each node also acts as a relay for transmitting data of
other connected nodes. Mesh topologies are further
divided into Fully Connected Mesh and Partially
Connected Mesh. In fully connected mesh topology,
each node is connected to every other node while in
partially connected mesh topology, a node is
connected one or more neighboring nodes.
21
Fig 7 / Network Topologies in WSN

22
 Applications of WSN
• Air Traffic Control (ATC)
• Heating Ventilation and Air Conditioning (HVAC)
• Industrial Assembly Line
• Automotive Sensors
• Battlefield Management and Surveillance
• Biomedical Applications
• Bridge and Highway Monitoring
• Disaster Management
• Earthquake Detection
• Electricity Load Management

23
• Environment Control and Monitoring
• Industrial Automation
• Inventory Management
• Personal Health Care
• Security Systems

24
 Topic 2

Single Node Architecture –

Hardware Components

25
 Introduction
• Building a wireless sensor network requires the
constituting nodes to be developed. These nodes have
to meet the requirements from a given application.
They have to be small, cheap, energy efficient,
equipped with the right sensors, memory resources
and sufficient communication facilities. The hardware
components of the functioning node are explained as
follows.

26
 Overview of Sensor Node
• A basic sensor node comprises five main components
are shown in the Figure.
• Controller: To process all relevant data
• Memory: To store programs and intermediate data.
• Sensors and actuators: Actual interface to the physical
world to observe or control physical parameters of the
environment.
• Communication: Device for sending and receiving
information over a wireless channel
• Power supply: Some form of batteries necessary to
provide energy and some form of recharging by
obtaining energy from the environment as well.
27
Fig 8 / Basic Components of a Sensor Node

28
 Controllers
• The controller is the core of a wireless sensor node.
• It is the Central Processing Unit (CPU) of the node
• It collects data from sensors, processes this data,
receives data from other sensor nodes, and decides on
the actuator’s behavior.
• It has to execute various programs, ranging from time-
critical signal processing and communication protocols
to application programs.
• Such a variety of processing tasks can be performed on
various controller architectures, representing trade-offs
between flexibility, performance, energy efficiency, and
costs.
29
• Microcontrollers are suitable for WSNs since they can
reduce their power consumption by going into sleep
states where only parts of the controller are active.
• One of the main differences to general-purpose
systems is that microcontroller-based systems do not
include a memory management unit – for example,
protected or virtual memory is difficult.
• In a wireless sensor node, DSP can be used to process
incoming data. But the advantages of a DSP are not
required in a WSN node and they are usually not used.
• Another option for the controller is to use Field-
Programmable Gate Arrays (FPGAs) or Application-
Specific Integrated Circuits (ASICs) instead of
microcontrollers. 30
• An FPGA can be reprogrammed in the field to adapt to
a changing set of requirements , but this can take time
and energy.
• An ASIC is a specialized processor, designed for a given
application such as high-speed routers and switches.
• The typical trade-off here is loss of flexibility in return
for a considerably better energy efficiency and
performance.

31
 Memory
• There is a need for Random Access Memory (RAM) to
store intermediate sensor readings, packets from other
nodes etc.
• RAM is fast, but it loses its contents if power supply is
interrupted.
• The program code can be stored in Read-Only Memory
(ROM) or in Electrically Erasable Programmable Read-
Only Memory (EEPROM) or flash memory.
• Flash memory can also serve as intermediate storage of
data when the power supply goes off for some time.
• The long read and write access delays of flash memory
should be taken into account as well as the high
required energy.
32
 Communication Module
1. Choice of transmission medium
• The first choice is the transmission medium and usual
choices include radio frequencies, optical
communication, and ultrasound.
• Radio Frequency (RF)-based communication is vital
requirement of most WSN applications.
• It provides long range and high data rates, acceptable
error rates at reasonable energy expenditure, and does
not require line of sight between sender and receiver.
• For a practical wireless, RF-based system, the carrier
frequency has to be carefully chosen. The wireless
sensor networks use communication frequencies
between about 433 MHz and 2.4 GHz.
33
 2. Transceivers
• For actual communication, both a transmitter and a
receiver are required in a sensor node to convert a bit
stream coming from a microcontroller and convert them
to and from radio waves. Such combined devices are
called transceivers.
• Usually, half-duplex operation is realized since
transmitting and receiving at the same time on a
wireless medium is impractical in most cases. A range of
low-cost transceivers is available that incorporate all the
circuitry required for transmitting and receiving,
modulation, demodulation, amplifiers, filters, mixers
etc..
34
 3. Transceiver tasks and characteristics
• The following are the some of the important
characteristics of a transceiver which should be taken
into account.
– Service to upper layer
– Power Consumption and Energy Efficiency
– Carrier Frequency & Multiple channels
– Transmission Power Control
– Data Rates
– Modulation
– Noise Figure
– Power Efficiency
– Frequency Stability etc
35
 4. Transceiver States
• Transmit State: The transmit part of the transceiver is
active and the antenna radiates energy.
• Receive State: The receive part is active.
• Idle State: A transceiver that is ready to receive but not
currently receiving anything is said to be in an idle
state.
• Sleep State: The significant parts of the transceiver are
switched off. There are transceivers offering several
different sleep states.

36
 Sensors & Actuators
• Sensors can be categorized into the following three
categories -
1. Passive Omni-directional sensors:
• They can measure a physical quantity at the point of
the sensor node without manipulating the environment
by active probing. They obtain the energy directly from
the environment – energy is only needed to amplify
their analog signal. There is no notion of “direction in
these measurements. Typical examples include
thermometer, light sensors, vibration, microphones,
humidity, chemical sensors etc

37
2. Passive narrow-beam sensors: They are passive but
have a well-defined notion of direction of
measurement. A typical example is a camera, which
can “take measurements” in a given direction, but has
to be rotated if need be.
3. Active sensors: They probe the environment, for
example, a sonar or radar sensor or some types of
seismic sensors, which generate shock waves by small
explosions.

38
 Power Supply of Sensor Nodes
1. Traditional batteries
• The power source of a sensor node is a battery, either
non-rechargeable (primary batteries) or, if an energy
scavenging device is present on the node, also
rechargeable (secondary batteries).
• In some form or other, batteries are electro-chemical
stores for energy – the chemicals being the main
determining factor of battery technology.

39
 2. Energy scavenging
• Some of the unconventional energy sources like fuel
cells, micro heat engines and radioactivity – convert
energy from stored secondary form into electricity in a
easy way than a normal battery would do.
• The entire energy supply is stored on the node itself –
once the fuel supply is exhausted, the node fails.
• The energy from a node’s environment must be tapped
into and made available to the node – energy
scavenging should take place.

40
3. Photo-voltaics
The solar cells can be used to power sensor nodes. The
available power depends on whether nodes are used
outdoors or indoors, and on time of day. The resulting
power ranges between 10 mW/cm2 indoors and 15
mW/cm2 outdoors. Single cells achieve a fairly stable
output voltage of about 0.6 V. Hence, solar cells are
used to recharge secondary batteries.
4. Temperature gradients
Differences in temperature can be directly converted to
electrical energy. Theoretically, even small difference
for example, 5 K can produce considerable power, but
practical devices fall very short of theoretical upper
limits.
41
5. Vibrations
Walls or windows in buildings are resonating with cars
or trucks passing in the streets, machinery often has
low- frequency vibrations, ventilations also cause it,
and so on. The available energy depends on amplitude
and frequency of the vibration and ranges between 0.1
mW/cm3 and 10, 000 mW/cm3 for some extreme
cases.

42
 Topic 3

Network Characteristics

43
• The following are the characteristics of Wireless Sensor
Networks:
1. Type of service
• The service type provided by a conventional
communication network is to move bits from one place
to another.. A WSN is expected to provide meaningful
information and actions about a given task. The
concepts like scoping of interactions to specific
geographic regions or to time intervals are important.
Hence using such a network along with new interfaces
and new ways of thinking about the service of a
network are required.

44
 2. Quality of Service
• The quality of service is closely related to the type of a
network’s service. The traditional quality of service
requirements coming from multimedia-type
applications like bounded delay or minimum
bandwidth are irrelevant when applications are
tolerant to latency or the bandwidth of the transmitted
data.
• In some cases, the occasional delivery of a packet can
be more than enough and in other cases, very high
reliability requirements exist. In some other cases,
delay is important when actuators are to be controlled
in a real-time fashion by the sensor network. The
packet delivery ratio is an insufficient metric.
45
• The vital requirement is the amount and quality of
information that can be extracted at given sinks about
the observed objects or area.
• Therefore, adapted quality concepts like reliable
detection of events or the approximation quality of a,
say, temperature map is important.

46
 3. Fault tolerance
• The nodes may run out of energy or get damaged, or
even interrupt the wireless communication between
two nodes permanently. The redundant deployment is
necessary for WSN to tolerate the node failure and
using more number of nodes will be necessary even if
all nodes functioned correctly.
4. Lifetime
• In many cases, the nodes will have to depend on a
limited supply of energy using batteries. Replacing
these energy sources in the field is usually not
practicable and simultaneously, a WSN must operate at
least for a given mission time. Hence, the lifetime of a
WSN becomes a very important figure of merit. Hence
an energy-efficient way of operation of the WSN is
necessary. As an alternative to energy supplies, a
limited power source must also be available on a
sensor node. 47
• These sources are not powerful enough to ensure
continuous operation but can provide some recharging
of batteries.
• Under such conditions, the lifetime of the network
should ideally be infinite. The lifetime of a network also
has direct trade-offs against quality of service: investing
more energy can increase quality but decrease lifetime.
• The precise definition of lifetime depends on the
application. The simple option is to use the time until
the first node fails as the network lifetime. Other
options include the time until the network is
disconnected in two or more partitions, the time until
50 % of nodes have failed etc.
48
 5. Scalability
• Since a WSN may include a large number of nodes,
the employed architectures and protocols must be
able scale to these numbers.
6. Wide range of densities
• In a WSN, the number of nodes per unit area that is
the density of the network can vary considerably.
Different applications will have very different node
densities. Even within a given application, density can
vary over time and space and density also does not
have to homogeneous in the entire and the network
should adapt to such variations.

49
 7. Programmability
▪ The nodes need to process information and react
flexibly on changes in their tasks. These nodes should
be programmable and their programming must be
changeable during operation when new tasks become
important.
8. Maintainability
• WSN has to monitor its own health and status to
change operational parameters or to choose different
trade-offs. The network can also be able to interact
with external maintenance mechanisms to ensure its
extended operation at a required quality.

50
 Topic 4

Challenges of WSN

51
 Introduction
• To realize the characteristics requirements, the
innovative mechanisms for a communication network
have to be found.
• The particular challenge is the need to find
mechanisms specific to the idiosyncrasies of a given
application to support the specific quality of service,
and maintainability requirements.
• These mechanisms also have to generalize to a wider
range of applications and implementation of a WSN
becomes necessary for every individual application.
• Some of the mechanisms that will form typical parts of
WSNs are:
52
 1. Multi-hop Wireless Communication
• Since wireless communication is a core technique, a
direct communication between a sender and a receiver
is faced with limitations.
• In particular, communication over long distances is only
possible using high transmission power.
• The use of intermediate nodes as relays can reduce the
total required power.
• Hence, for many forms of WSNs, multi-hop
communication will be a necessary ingredient.

53
 Energy Efficient Operation & Auto-configuration
• 2. Energy-efficient Operation: It is a key technique for
supporting long life time. The other options include
energy-efficient data transport between two nodes or
the energy-efficient determination of requested
information. The non-homogeneous energy
consumption – the forming of “hotspots” is an issue.
• 3. Auto-configuration: A WSN will have to configure
most of its operational parameters, independent of
external configuration. As an example, nodes should
be able to determine their geographical positions only
using other nodes of the network so- called “self-
location”. The network should be able to tolerate failing
nodes or to integrate new nodes.
54
 4. Collaboration & In-network Processing
• In some applications, a single sensor is not able to decide
whether an event has happened but several sensors have
to collaborate to detect an event and only the joint data of
many sensors provides enough information.
• Information is processed in the network in various forms
to achieve this collaboration. This is opposite to having
every node transmit all data to an external network and
process it “at the edge” of the network.
• An example is to determine the highest or the average
temperature within an area and to report that value to a
sink. To solve such tasks, readings from individual sensors
can be aggregated reducing the amount of data to be
transmitted and hence improving the energy efficiency.
55
 5. Data Centric
• Traditional communication networks are centered
around the transfer of data between two specific
devices, each equipped with one network address –
the operation of such networks is thus address-centric.
• In a WSN, the nodes are deployed to protect against
node failures or to compensate for the low quality of a
single node’s actual sensing equipment. Hence,
switching from an address-centric paradigm to a data-
centric paradigm in designing architecture and
communication protocols is promising.
• An example for such a data-centric interaction will be
to request the average temperature in a given location
area, as opposed to requiring temperature readings
from individual nodes.
56
 6. Locality
• The principle of locality will have to be embraced to
ensure in particular, scalability.
• Nodes with limited should attempt to limit the state
that they accumulate during protocol processing to
only information about their direct neighbors.
• This will allow the network to scale to large numbers of
nodes without having to depend on powerful
processing at each single node.

57
 7. Exploit Trade-offs
• Similar to locality principle, WSNs will have to depend
to a large degree on exploiting various trade-offs
between contradictory goals, both during system
design and runtime.
• Examples for such trade-offs are - higher energy
expenditure allows higher result accuracy, longer
lifetime of the entire network trades off against
lifetime of individual nodes and node density.
• If there is a depart from an address-centric view of the
network, it may require new programming interfaces
beyond the simple semantics of the conventional
socket interface and allow concepts like required
accuracy, energy/accuracy trade-offs etc.
58
 Topic 5

Enabling Technologies for WSN

59
 Introduction
• It has only become possible to build wireless sensor
networks with some fundamental advances in enabling
technologies.
• First and foremost among these technologies is the
miniaturization of hardware.
• Smaller feature sizes in chips have driven down the
power consumption of the basic components of a sensor
node to a level that the constructions of WSNs can be
contemplated.
• This is particularly relevant to microcontrollers and
memory chips and the radio modems responsible for
wireless communication.
60
• Reduced chip size and improved energy efficiency is
accompanied by reduced cost, which is necessary to
make redundant deployment of nodes affordable.
• The actual sensing equipment is the third relevant
technology next to processing and communication.
• However, it is difficult to generalize because of the vast
range of possible sensors.

61
Fig 9 / Enabling Technologies for WSN

62
 Energy Scavenging
• These three basic parts of a sensor node have to be
accompanied by power supply.
• This requirement depends on application, high capacity
batteries lasting for long times and can efficiently provide
small amounts of current.
• A sensor node also has a device for energy scavenging,
recharging the battery with energy gathered from the
environment – solar cells or vibration-based power
generation are conceivable options.
• Such a concept requires the battery to be efficiently
chargeable with small amounts of current, which is not a
standard ability.
• The counterpart to the basic hardware technologies is
software.
63
• The architecture of the operating system or runtime
environment has to support simple re-tasking, cross-
layer information exchange and modularity to allow for
simple maintenance.
• This software architecture on a single node has to be
extended to a network architecture, where the division
of tasks between nodes is considered.
• The third part to solve is how to design appropriate
communication protocols.
• Figure 9 shows the enabling technologies for WSN.

64
 Topic 6

Types of Wireless Sensor Networks

65
 Introduction
• The types of networks are decided based upon the
environment so that they can be deployed underwater,
underground, on land and so on. Different types of
WSNs include:
– Terrestrial WSNs
– Underground WSNs
– Underwater WSNs
– Multimedia WSNs
– Mobile WSNs

66
 Terrestrial WSN’s
• Terrestrial WSNs are capable of communicating base
stations efficiently and consist of hundreds to
thousands of wireless sensor nodes deployed either in
an unstructured or structured manner.
• In an unstructured mode, the sensor nodes are
randomly distributed within the target area dropped
from a fixed plane.
• The preplanned or structured mode considers optimal
placement, grid placement, and 2D, 3D placement
models. In this WSN, the battery power is limited but
equipped with solar cells as a secondary power source.
• The energy conservation of these WSNs is achieved by
using low duty cycle operations, minimizing delays, and
optimal routing, and so on.
67
68
 Underground WSN
• The underground wireless sensor networks are more
expensive than the terrestrial WSNs in terms of
deployment, maintenance, and equipment cost
considerations and careful planning.
• The WSNs networks consist of several sensor nodes
hidden in the ground to monitor underground
conditions.
• To relay information from the sensor nodes to the base
station, additional sink nodes are located above the
ground.
• The underground wireless sensor networks deployed
into the ground are difficult to recharge.
69
• The sensor battery nodes equipped with limited
battery power are difficult to recharge
• In addition to this, the underground environment
makes wireless communication a challenge due to the
high level of attenuation and signal loss.

70
Fig 10 / Underground WSN

71
 Under Water WSN
• More than 70% of the earth is occupied with water.
These networks consist of several sensor nodes and
vehicles deployed underwater.
• Autonomous underwater vehicles are used for
gathering data from these sensor nodes. A challenge of
underwater communication is a long propagation
delay, and bandwidth and sensor failures.
• Underwater, WSNs are equipped with a limited battery
that cannot be recharged or replaced.
• The issue of energy conservation for underwater WSNs
involves the development of underwater
communication and networking techniques.
72
Fig 11 / Underwater WSN

73
 Multimedia WSN
• Multimedia wireless sensor networks have been proposed
to enable tracking and monitoring of events in the form of
multimedia such as imaging, video, and audio.
• These networks consist of low-cost sensor nodes equipped
with microphones and cameras.
• These nodes are interconnected with each other over a
wireless connection for data compression, data retrieval,
and correlation.
• The challenges with the multimedia WSN include high
energy consumption, high bandwidth requirements, data
processing, and compressing techniques.
• In addition to this, multimedia contents require high
bandwidth for the content to be delivered properly and
easily.
74
Fig 12 / Multimedia WSN

75
 Mobile WSN
• These networks consist of a collection of sensor nodes
that can be moved on their own and can be interacted
with the physical environment.
• The mobile nodes can compute sense and
communicate. Mobile wireless sensor networks are
much more versatile than static sensor networks.
• The advantages of MWSN over static wireless sensor
networks include better and improved coverage, better
energy efficiency, superior channel capacity, and so on.

76
77
 Classification of WSN’s
• The classification of WSNs can be done based on the
application but its characteristics mainly change based
on the type.
• Generally, WSNs are classified into different categories
like the following.
– Static & Mobile
– Deterministic & Nondeterministic
– Single Base Station & Multi Base Station
– Static Base Station & Mobile Base Station
– Single-hop & Multi-hop WSN
– Self Reconfigurable & Non-Self Configurable
– Homogeneous & Heterogeneous
78
 1. Static & Mobile WSN
• All the sensor nodes in several applications can be set
without movement so these networks are static WSNs.
Especially in some applications like biological systems uses
mobile sensor nodes which are called mobile networks.
The best example of a mobile network is the monitoring of
animals.
2. Deterministic & Nondeterministic WSN
• In a deterministic type of network, the sensor node
arrangement can be fixed and calculated. This sensor
node’s pre-planned operation is possible in simply some
applications. In most applications, the location of sensor
nodes cannot be determined because of different factors
like hostile operating conditions and harsh environment,
so these networks are called non-deterministic.
79
 3. Single Base Station & Multi Base Station
• In a single base station network, a single base station is
used and it can be arranged very close to the region of
the sensor node. The interaction between sensor nodes
can be done through the base station. In a multi-base
station type network, multiple base stations are used
and a sensor node is used to move data toward the
nearby base station.
4. Static Base Station & Mobile Base Station
• Base stations are either mobile or static similar to
sensor nodes. The static type base station includes a
stable position close to the sensing area whereas the
mobile base station moves in the region of the sensor
so that the sensor nodes load can be balanced.
80
 5. Single-hop & Multi-hop WSN
• In a single-hop type network, the arrangement of
sensor nodes can be done directly toward the base
station whereas, in a multi-hop network, both the
cluster heads and peer nodes are utilized to transmit
the data to reduce the energy consumption.
6. Self Reconfigurable & Non-Self Configurable
• In a non-self configurable network, the arrangement of
sensor networks cannot be done by them within a
network and depends on a control unit for gathering
data. In wireless sensor networks, the sensor nodes
maintain and organize the network and collaboratively
work by using other sensor nodes to accomplish the
task.
81
 7. Homogeneous and Heterogeneous
• In a homogeneous wireless sensor network, all the
sensor nodes mainly include similar energy utilization,
storage capabilities and computational power.
• In heterogeneous network, some sensor nodes include
high computational power as well as energy necessities
as compared to others.
• The processing and communication tasks are separated
consequently.

82
Model Question Bank

83
 PART A
1. What is a sensor?
2. What is a sensor network?
3. Give the elements of WSN.
4. What are the basic components of a sensor node?
5. Differentiate between single hop and multi-hop
networks.
6. Differentiate between flat and hierarchical network
architectures.
7. What are the various topologies used in WSN?
8. Give any four applications of WSN.
9. How does wireless sensor network work?
10. What is the need for wireless sensor network?
84
11. Mention the challenges of wireless sensor networks.
12. What is event detection?
13. What is energy scavenging?
14. Differentiate between sensor and actuator.
15. What is quality of service?
16. List the types of WSN.
17. What is Multi-hop wireless communication?
18. What is data centric?
19. What is an active sensor?
20. State the deployment options.

85
 PART B
1. Discuss briefly the various hardware components used
in Single node architecture of WSN.
2. Explain the characteristics, constraints and challenges
of WSN.
3. Write a short note on enabling technologies for
wireless sensor networks.
4. Describe the types of wireless sensor networks in a
brief manner.

86
Wireless Sensor Networks

Unit 2 / Architectures

Prepared By

Dr.S.Omkumar/Associate Prof

1
 Syllabus / Unit 2
• Network Architecture- Sensor Networks-
Scenarios- Design Principle, Physical Layer and
Transceiver Design Considerations, Optimization
Goals and Figures of Merit, Gateway Concepts,
Operating Systems and Execution Environments-
Introduction to TinyOS and nesC- Internet to
WSN Communication.

2
 Topic 1

Sensor Networks Scenario

3
 Types of Sources & Sinks
• A source is any entity in the network that provide
information typically a sensor node and also be an
actuator node that provides feedback about an
operation.
• A sink is the entity where information is required.
• There are three options for a sink - it can belong to the
sensor network or just another sensor/actuator node or
can be an entity outside this network.
• For the second case, the sink can be an actual device to
interact with the sensor network or can also be a
gateway to another larger network such as Internet.
• These main types of sinks are shown in Figure 2.1,
showing sources and sinks in direct communication.
4
Fig 1 / Three Types of Sinks

5
 Single Hop Vs Multiple Hop Networks
• Because of limited distance, the simple direct
communication between source and sink is not possible
in WSN, which are intended to cover a lot
environmental or agriculture applications.
• To overcome such limited distances, the relay stations
are used with the data packets taking multi hops from
the source to the sink.
• The multi-hopping is a working solution to overcome
problems with large distances and can also improve the
energy efficiency of communication.
• It consumes less energy to use relays instead of direct
communication.
6
• The energy is actually wasted if intermediate relays are
used for short distances and for large distance, the
radiated energy dominate the fixed energy costs
consumed in transmitter and receiver electronics.
• Moreover multi-hop networks operate in a store and
forward fashion.

7
Fig 2 / Multihop Network

8
 Multiple Sinks & Sources
• So far, only networks with a single source and a single
sink have been explained.
• In many cases, there are multiple sources and/or
multiple sinks present. In the most challenging case,
multiple sources should send information to multiple
sinks, where either all or some of the information has to
reach all or some of the sinks. Figure 2.3 illustrates
these combinations.

9
 Types of Mobility
• One of the main virtues of wireless communication is its
ability to support mobile participants. In wireless sensor
networks, mobility can appear in three main forms:
1. Node mobility:
The wireless sensor nodes can be mobile. The meaning
of such mobility is highly application dependent. In node
mobility, the network has to reorganize itself frequently
enough to be able to function correctly. There are trade-
offs between the frequency and speed of node
movement on one hand and the energy required to
maintain a desired level of functionality in the network
on the other hand.
10
2. Sink mobility:
The information sinks can be mobile. The important
aspect is the mobility of an information sink that is not
part of the sensor network, for example, a human user
requested information via a PDA while walking in an
intelligent building. In a simple case, such a requester
can interact with the WSN at one point and complete its
interactions before moving on. In many cases,
consecutive interactions can be treated as separate
unrelated requests.

11
Fig 4 / Mobile Sink through Sensor Network

12
 3. Event mobility:
• In applications like event detection and in tracking, the
cause of the events or the objects to be tracked can be
mobile. In such scenarios, the observed event is covered
by a sufficient number of sensors at all time.
• Hence, sensors will wake up around the object, engaged
in higher activity to observe the present object, and
then go back to sleep. As the event source moves
through the network, it is accompanied by an area of
activity within the network. This is called as Frisbee
Model as shown in Figure 2.4

13
 Topic 2

Design Principles for WSN

14
 Distributed Organization
• Both the scalability and the robustness optimization
goal are required to organize the network in a
distributed fashion.
• When organizing a network in a distributed fashion, it is
necessary to know potential shortcomings of this
approach.
• In many cases, a centralized approach can produce
solutions that perform better or require fewer
resources.
• One possibility is to use centralized principles in a
localized fashion by electing, out of set of equal nodes.
• Such elections result in a dynamic hierarchy.
• The election process should be repeated continuously
until the elected node runs out of energy 15
 In Network Processing Techniques
1. Aggregation:
• The simplest in-network processing technique is
aggregation. The term aggregation means that
information is aggregated into a condensed form in
nodes intermediate between sources and sinks out of
information provided by nodes further away from the
sink. The aggregation function must be applied in the
intermediate nodes as shown in Figure 2.5.

16
Fig 5 / Aggregation as an Example

17
 2. Distributed Source Coding and Distributed
Compression:
• The objective is to encode the information provided by
several sensors by using traditional coding schemes,
which may be complex for simple sensor nodes.
• The readings of adjacent sensors are going to be quite
similar and correlated.
• Such correlation can be exploited instead of sending the
sum of the data so that the overhead can be reduced.

18
 3. Distributed and collaborative signal processing
• When complex computations on a certain amount of
data is to be done, it can be more energy efficient to
compute these functions on the sensor nodes using Fast
Fourier Transform (FFT). In principle, this is similar to
algorithm design for parallel computers. However the
energy consumption of communication and
computation are relevant parameters to decide
between various algorithms.
4. Mobile code/Agent-based networking
• The idea of mobile code is to have a small, compact
representation of program code to be sent from node to
node. This code is executed locally for collecting
measurements and then decides where to be sent next.
This idea has been used in various environments
19
 Adaptive Fidelity & Accuracy
• The idea of making fidelity of computation depends
upon the amount of energy available for that particular
computation.
• This concept can be extended from a single node to an
entire network. As an example, consider a function
approximation application.
• When more sensors participate in the approximation,
the function is sampled at more points and the
approximation is better. But more energy has to be
invested.
• Hence, it is up to an application to define the degree of
accuracy of the results and the task of the
communication protocols to achieve this accuracy. 20
 Data Eccentricity
• In traditional communication networks, the focus will be
on the pair of communicating peers, the sender and the
receiver of data.
• In a wireless sensor network, the interest of an
application is actual information reported about the
physical environment. This is applicable when a WSN is
redundantly deployed such that any given event can be
reported by multiple nodes.
• This method of concentrating on the data rather than
identity of nodes is called data-centric networking.
• For an application, this means that an interface is
exposed by the network where data only is addressed in
requests. 21
 Exploit Local Information
• Another useful technique is to exploit location
information in the communication protocols when-ever
such information is present.
• Since the location of an event is crucial information for
many applications, mechanisms must be available to
determine the location of sensor nodes.
• It can simplify the design and operation of
communication protocols and can improve their energy
efficiency.

22
 Exploit Activity Patterns
• Activity patterns in a wireless sensor network are quite
different from that of traditional networks.
• The data rate averaged over a long time can be very
small.
• This can be detected by a larger number of sensors,
breaking into a frenzy of activity, causing a well-known
event shower effect.
• Hence, the protocol design should be able to handle
such bursts of traffic by switching between modes of
quiescence and of high activity.

23
 Exploit Heterogeneity
• Sensor nodes can be heterogeneous by constructions,
that is, they have larger batteries, farther-reaching
communication devices, or more processing power.
• They can also be heterogeneous by evolution, that is,
they started from an equal state, but scavenge energy
from the environment due to overloading.
• Heterogeneity in the network is both a burden and an
opportunity.
• The opportunity is an asymmetric assignment of tasks,
giving nodes with more resources or more capabilities
the more demanding tasks.
• The burden is asymmetric task assignments cannot be
static but have to be reevaluated.
24
 Component Based Protocol Stacks
• The concept is a collection of components which can form
a basic “toolbox” of protocols and algorithms to build
upon.
• All wireless sensor networks will require some form of
physical, MAC, Link layer protocols, routing and transport
layer functionalities.
• Moreover, “helper modules” like time synchronization,
topology control can be useful.
• On top of these basic components, more abstract
functionalities can then be built.
• The set of components active on a sensor node can be
complex and will change from application to application.
• Protocol components will also interact with each other
either by using simple exchange of data packets or by
exchange of cross-layer information.
25
 Topic 3

Physical Layer and Transceiver

Considerations

26
 Introduction
• Some of the crucial points influencing the Physical Layer
design in wireless sensor networks are -
– Low power consumption
– Small transmit power and a small transmission range
– Low duty cycle
– Low data rates in the order of tens to hundreds
kilobits per second
– Low implementation complexity and costs
– Low degree of mobility
– Small form factor for the overall node

27
 Energy Usage Profile
• The choice of a small transmission power leads to an
energy consumption profile different from other
wireless devices like cell phones.
• The radiated energy is small and the overall transceiver
consumes much more energy than actually radiated.
• Then for small transmit powers, transmit and receive
modes consume more or less the same power
depending on the transceiver architecture.
• To reduce average power consumption in a low-traffic
wireless sensor network, the transceiver must go into
sleep state instead of just idling.
• During this startup time, no transmission or reception of
data is possible.
28
• The third key observation is the relative costs of
communications versus computation in a sensor node.
• A comparison of these costs depends for the
communication part on BER requirements, range,
transceiver type etc.

29
 Choice of Modulation Scheme
• The following factors have to be balanced for the choice
of modulation scheme -
– Required data rate and symbol rate
– Implementation complexity
– Relationship between radiated power and target BER
– Expected channel characteristics
• To maximize the time of transceiver in sleep mode, the
transmit times should be minimized. The higher the
data rate offered by a modulation, the smaller the time
needed to transmit a given amount of data and the
smaller the energy consumption. Moreover, the power
consumption of a modulation scheme depends much
more on the symbol rate than on the data rate.
30
 Dynamic Modulation Scaling
• To adapt the modulation scheme to the current
situation, an approach called dynamic modulation
scaling is employed.
• For the case of m-ary QAM, a model has been
developed with the symbol rate ‘B’ and the number of
levels per symbol ‘m’ as parameters.
• This model expresses the energy required per bit and
also the achieved delay per bit, taking into account the
higher levels of modulation.
• Hence the bit delay decreases for increasing values of
‘B’ and ‘m’. The energy per bit depends much more on
‘m’ than on ‘B’.
31
• For the particular parameters chosen, both energy per
bit and delay per bit can be minimized for the maximum
symbol rate.
• With modulation scaling, a packet is equipped with a
delay constraint, from which directly a minimum
required data rate can be derived.

32
 Antenna Considerations
• The desired small form factor of the overall sensor
nodes restricts the size and the number of antennas.
• If the antenna is much smaller than the carrier’s
wavelength, it is difficult to achieve good antenna
efficiency.
• In case of small sensor node cases, it will be difficult to
place two antennas with suitable distance to achieve
receive diversity.
• The antennas should be spaced apart at least 40–50% of
the wavelength used to achieve good effects from
diversity.

33
• In addition, radio waves emitted from an antenna close
to the ground are faced with higher path-loss
coefficients than the common value α = 2 for free-space
communication.
• Moreover, depending on the application, antennas must
not protrude from the casing of a node to avoid possible
damage to it.
• These restrictions limit the quality and characteristics of
an antenna for wireless sensor nodes.

34
 Topic 4

Optimization of Goals & Figure of

Merit

35
 Introduction
• The following techniques will optimize a
network, compare solutions, decide a better
approach for a given application, and turn
optimization goals into measurable figures of
merit.

36
 Quality of Service
• WSNs differ from other conventional communication
networks in the type of services they offer.
• These networks only move bits from one place to
another.
• Such QoS can be regarded as a low-level, networking-
device attributes like bandwidth, delay, jitter or as a
high-level, user attributes like perceived quality of a
voice communication or a video transmission.
• But high-level QoS attributes in WSN highly depend on
the application.
• Some generic possibilities are:

37
1. Event detection/reporting probability
The probability of an event that actually occurred is
not detected or not reported to an information sink
2. Event classification error
If events are to be both detected and classified, the
error in classification must be small.
3. Event detection delay
The delay between detecting an event and reporting to
all interested sinks
4. Missing reports
The probability of undelivered reports should be small
in periodic reporting applications.
38
5. Approximation accuracy
For function approximation applications, the
average/maximum absolute error with respect to the
actual function.
6. Tracking accuracy
In Tracking applications, the reported position should
be as close to the real position and the error should be
small.

39
 Energy Efficiency
• The most commonly considered aspects of energy
efficiency are:
1. Energy per correctly received bit
The average amount of energy to transport one bit of
information from the source to the destination.
2. Energy per reported event
The average energy spent to report one event
3. Delay/energy trade-offs
The notion of “urgent” events to justify the increased
energy investment for a speedy reporting of events.

40
4. Network lifetime
• The time for which the network is operational to fulfill
its tasks starting from a given amount of stored energy.
– Time to first node death: First node in the network run out of
energy and stop operating
– Network half-life: When 50% of the nodes run out of energy
and stopped operating.
– Time to partition: First partition of the network in two or
more disconnected parts occur
– Time to Loss of Coverage: For the first time any spot in the
deployment region is no longer covered by any node’s
observations.
– Time to failure of first event notification: The unreachable
part of the network does not want to report any events in the
first place.
41
 Scalability
• The ability to maintain performance characteristics
irrespective of the size of the network is called
scalability.
• Scalability requires consistent state such as addresses or
routing table entries to be maintained.
• Hence, the need to restrict such information is enforced
with the resource limitations of sensor nodes with
respect to memory.
• The need for extreme scalability has direct
consequences on the protocol design.
• Architectures and protocols should implement
appropriate scalability support rather than trying to be
as scalable as possible.
42
 Robustness
• Related to QoS and scalability requirements, wireless
sensor networks should also exhibit an appropriate
robustness.
• They should not fail just because a limited number of
nodes run out of energy, or because their environment
changes.
• These failures have to be compensated by finding other
routes.
• A precise evaluation of robustness is difficult in practice
and depends mostly on failure models for both nodes
and communication links

43
 Topic 5

Gateway Concepts

44
 Need for Gateways
• For practical deployment, the sensor network has to
interact with other information devices. The standard
example is to read the temperature sensors in one’s
home while traveling. Figure 2.6 shows the networking
scenario.
• The WSN has to exchange data with such a mobile
device or with some sort of gateway which provides the
physical connection to the Internet.
• The first option is to regard a gateway as a simple router
between Internet and sensor network. This will entail
the use of Internet protocols within the sensor network.
• The next option is to design the gateway as an actual
application-level gateway on the basis of the
application-level information.
45
Fig 6 / WSN with Gateway Node

46
 WSN to Internet Communication
• For example, a sensor node wants to deliver an alarm
message to some Internet host.
• The first problem to solve is to find the gateway from
within the network.
• If several gateways are available, the selection of the
particular route and gateway for a given destination have
to be done.
• To handle several gateways the option is to build an IP
overlay network on top of the sensor network. Figure 2.7
shows the mapping of Alice to a concrete IP address.
• The sensor node has to include sufficient information such
as IP address and port number in its own packets.
• The gateway in turn will extract this information and
translate it into IP packets.
47
Fig 7 / WSN to Internet Communication

48
 Internet to WSN Communication
• For example, a mobile requester is equipped with a
WSN transceiver which has all the necessary protocol
components.
• In this case, the requesting terminal can be a direct part
of the WSN.
• First of all, identification of the sensor network in the
desired location and existence of a gateway node has to
be done.
• Once the requesting terminal has obtained this
information, then the actual services can be accessed.
• The requesting terminal can instead send a properly
formatted request to this gateway which acts as an
application-level gateway that can answer this request.
49
• The gateway translates this request into the proper intra
sensor network protocol interactions. Figure 8 shows
the scenario.

50
 WSN Tunneling
• The gateways can also act as simple extensions of one
WSN to another WSN.
• The idea is to build a larger virtual WSN “tunneling” all
protocol messages between these two networks and
simply using the Internet as a transport network as
shown in Figure 2.9.
• But care has to be taken not to confuse the virtual link
between two gateway nodes with a real link.
• Otherwise, protocols that depend on physical properties
of a communication link can get confused.

51
Fig 9 / WSN Tunneling

52
 Topic 6

Operating Systems & Execution

Environment

53
 Embedded Operating Systems
• The traditional tasks of an operating system are controlling
and protecting the access to resources, managing their
allocation to users and support for concurrent execution of
processes.
• These tasks are only partially required in an embedded
system and these systems do not have required resources
to support a full-blown operating system.
• In particular, the need for energy-efficient execution
requires support for energy management or Dynamic
Voltage Scaling (DVS) techniques.
• Also, external components like sensors, the radio modem,
or timers should be handled easily and efficiently.
• All this requires an appropriate programming model to
structure a protocol stack and explicit support for energy
management. 54
 Programming Paradigms
1. Concurrent Programming
• The support for concurrent execution is crucial for WSN
nodes to handle data coming from arbitrary sources like
multiple sensors or the radio transceiver at arbitrary
points in time.
• For example, a system can poll a sensor to decide
whether data is available and process the data, then poll
the transceiver to check whether a packet is available
and then immediately process the packet and so on.

55
2. Process Based Concurrency
• Most general-purpose operating systems support
concurrent execution of multiple processes on a single
CPU. Hence such a process-based approach can be used
to support concurrency in a sensor node as illustrated in
(b) of Figure 2.10.
• Mapping such an execution model of concurrent
processes to a sensor node shows that there are some
granularity mismatches.
• This problem is severe for smaller tasks to be executed
when compared to overhead.

56
Fig 10 / Programming Models for WSN

57
3. Event-based Programming
• The system waits for any event to happen, where an
event can be the availability of data from a sensor, or
arrival of a packet.
• Such an event is then handled by a short sequence of
instructions that stores the occurrence of event and
necessary information.
• This is called event based programming model as
shown in Figure 2.11.
• This programming model distinguishes between two
different “contexts”: - time-critical event handlers
(execution cannot be interrupted) and for the
processing of normal code (only triggered by the event
handlers).
58
Fig 11 / Event Based Programming Model

59
4. Interfaces to Operating System
• In WSNs, the interfaces should be accessible from
protocol implementations.
• This interface is closely tied with the structure of
protocol stacks.
• For example Application Programming Interface (API)
comprises, a “functional interface, object abstractions,
and detailed behavioral semantics”.
• Abstractions are wireless links, nodes and so on.
• The possible functions include state inquiry,
manipulation, transmitting of data, access to hardware
and setting of policies.

60
 Operating System & Protocols Stack
• In communication protocol structuring, the individual
protocols are stacked on top of each other, each layer only
using functions of the layer directly below.
• This layered approach has multiple benefits in keeping the
entire protocol stack manageable.
• As an example, consider the use of information about the
strength of the signal received from a communication
partner.
• This physical layer information can be used to assist in
networking protocols to decide about routing changes.
• Hence, one single source of information can be used by
many other protocols not directly associated with the
source of this information.
• Such cross-layer information exchange is one way to
loosen the strict confinements of the layered approach. 61
 Dynamic Energy & Power Management
1. Probabilistic State Transition Policies
• These policies regulate the transition between various
sleep states.
• They start out by considering sensors randomly
distributed over a fixed area and events arrive with
certain temporal distributions and spatial distributions.
• This allows them to compute probabilities for the time
to the next event, once an event has been processed.

62
2. Controlling Dynamic Voltage Scaling
• For example, only a single task has to be run in an
operating system. Hence, a clever scheduler is required
to decide exact clock rate to use in that situation to
meet all deadlines. This can require feedback from
applications for example, video playback in reference.
3. Trading off fidelity against energy consumption
• There are certain tasks that can be computed with a
higher or lower level of accuracy. The fidelity achieved by
such tasks is a candidate for trading off against other
resources. In a WSN, the natural trade-off is against
energy required to compute a task.

63
 Topic 7

Introduction to TinyOS & NesC

64
 Introduction
• The event-based programming model is the only
feasible way to support the concurrency required for
sensor node software with simple hardware provided by
these nodes.
• In addition, modularity should be supported to easily
exchange one state machine against another.
• The operating system TinyOS along with the
programming language nesC addresses these challenges
as follows.

65
 Events & Commands
• TinyOS supports modularity and event-based
programming by the concept of components.
• A component contains related functionality, for example,
for handling a radio interface.
• Such a component comprises the required state
information in a frame, program code for normal tasks and
handlers for events & commands.
• Both events and commands are exchanged between
different components.
• Components are arranged from low-level components
close to the hardware to high-level components making
up the actual application.
• Events originate in the hardware and pass upward from
low-level to high-level components whereas commands
are passed from high-level to low-level components. 66
 Event Handlers
• Figure 2.12 shows a timer component. It understands
three commands (“init”, “start”, and “stop”) and can
handle one event (“fire”) from another component.
• It issues “setRate” commands to this component and can
emit a “fired” event.
• In event-based paradigm, both command and event
handlers must run to conclusion and supposed to perform
very simple triggering duties.
• Commands must not block or wait for an indeterminate
amount of time.
• Similarly, an event handler only leaves information in its
component’s frame and arranges for a task to be executed
later.
• The actual computational work is done in the tasks.
• In TinyOS, they have to run to completion, but can be
interrupted by handlers.
67
Fig 12 / Example Timer Component

68
 FIFO Scheduler
• There are two advantages - there is no need for stack
management and tasks are atomic with respect to each
other.
• The arbitration between tasks can be triggered by
several events and are ready to execute.
• This is done by a simple First In First Out (FIFO)
scheduler, which shuts the node down when there is no
task executing or waiting.

69
 Split Phase Programming
• The first phase is the sending of the command and the
second is explicit information about the outcome of the
operation, delivered by a separate event.
• This split-phase programming approach requires for each
command a matching event that enables concurrency.
• When using split-phase programming, a large number of
commands and events are combined in a large program.
Hence, an abstraction is required to organize them.
• The set of commands that a component understands and
the set of events that a component may emit are its
interface to the components of a hierarchically higher
layer.
• Therefore, structuring commands and events forms an
interface between two components.
70
 NesC Language
• The nesC language allows a programmer to define
interface types that define commands and events
belong together.
• This allows split-phase programming style to put
commands and their corresponding completion events
into the same interface.
• Components then provide certain interfaces to their
users and in turn use other interfaces from underlying
components.
• Figure 2.13 shows the Timer component reorganized
into using a clock interface and providing two interfaces
StdCtrl and Timer.
71
Fig 13 / Timer Component using Interfaces

72
 Timer Component
• The TimerComponent is defined as a primitive
component containing handlers and tasks.
• Such primitive components or modules can be
combined into larger configurations by simply “wiring”
appropriate interfaces together.
• For this wiring only components that have the correct
interface types can be plugged together.
• Figure 2.14 shows how the TimerComponent and an
additional component HW Clock can be wired together
to form a new component Complete Timer.

73
Fig 14 / Larger Configuration

74
Model Question Bank

75
 PART A
1. Differentiate between a source and sink
2. Mention the three options for a sink.
3. What is multiple sink?
4. What is multi-hopping?
5. Give the types of mobility in WSN.
6. What is event mobility?
7. What is sink mobility?
8. Mention the techniques used for in-network
processing.
9. What is aggregation?
10. What is the use of mobile code?
76
11. What is meant by data centric networking?
12. Mention any ‘4’ crucial points influencing design of
physical layer in WSN.
13. Give the factors to be balanced for the choice of
modulation scheme.
14. Define figure of merit.
15. What is dynamic modulation scaling?
16. What are the various aspects of Energy efficiency?
17. What is scalability?
18. What are gateway concepts?
19. What is called tunneling?
20. What is concurrent programming?
21. What is TinyOS?
22. What is the use of NesC language?
77
 PART B
1. Explain the various scenarios of Sensor Networks.
2. Discuss in detail, the design principles of WSN.
3. Describe about optimization goals of a WSN and
figures of merit in detail.
4. Write a short note on Gateway Concepts.
5. Discuss in detail the characteristics and structure of
Transceivers.
6. Write a short note on TinyOS and NesC.

78
Wireless Sensor Networks

Unit 3 / Networking Sensors

Prepared
By
Dr.S.Omkumar

1
 Syllabus / Unit 3
• NETWORKING SENSORS:
• MAC Protocols for Wireless Sensor Networks,
Low Duty Cycle Protocols And Wakeup Concepts
– SMAC - B-MAC Protocol, IEEE 802.15.4
standard and ZigBee, the Mediation Device
Protocol, Wakeup Radio Concepts, Address and
Name Management, Assignment of MAC
Addresses, Routing Protocols Energy-Efficient
Routing, Geographic Routing

2
 Topic 1

Fundamentals of MAC Protocols

3
 Introduction to MAC Protocols
• The MAC protocol determines the points in time to
transmit a data, control or manage packet to another
node (unicast) or to a set of nodes (multicast,
broadcast).
• Two important responsibilities of the DLL are error
control and flow control.
• Error control is used to ensure correctness of
transmission and take appropriate actions in case of
transmission errors and flow control regulates the rate
of transmission.

4
• The important performance requirements for MAC
protocols are throughput efficiency, stability, fairness,
low access, low transmission delay and low overhead.
• The overhead can result from per-packet overhead
collisions, or from exchange of extra control packets.
• Collisions can happen if MAC protocol allows two or
more nodes to send packets at the same time.
• Collisions can result in the inability of the receiver to
decode a packet correctly, causing the upper layers to
perform a retransmission.

5
 Hidden Terminal Scenario
• If two nodes are out of reach, they cannot hear each
other. This gives rise to the hidden-terminal problems.
• The hidden-terminal problem occurs for Carrier Sense
Multiple Access (CSMA) protocols, where a node senses
the medium before starting to transmit a packet.
• If the medium is found to be busy, the node defers its
packet to avoid a collision and a subsequent
retransmission.

6
• Consider the example shown in Figure 3.1.
– Three nodes A, B, C arranged such that A and B are in
mutual range, B and C are in mutual range, but A and
C cannot hear each other.
– Now A starts to transmit a packet to B and sometime
later node C also decides to start a packet
transmission.
– A carrier-sensing operation by C shows an idle
medium since C cannot hear A’s signals.
– When C starts its packet, the signals collide at B and
both packets are useless.
– Using simple CSMA in a hidden-terminal scenario
thus leads to needless collisions.
7
 Exposed Terminal Scenario
– B transmits a packet to A and some moment later, C
wants to transmit a packet to D.
– This will be possible since both A and D will receive their
packets without distortions.
– The carrier-sense operation performed by C suppresses
C’s transmission and bandwidth is wasted.
• Using simple CSMA in an exposed terminal scenario
thus leads to needless waiting.
• Two solutions to the hidden-terminal and exposed-
terminal problems are busy-tone solutions and the
RTS/CTS handshake used in the IEEE 802.11 WLAN
standard.
8
 Classes of MAC Protocols
1. Fixed Assignment Protocols:
• The available resources are divided between the
nodes such that resource assignment is long term
without the risk of collisions.
• Long term means that the assignment is for durations
of minutes, hours, or even longer.
• To account for changes in topology due to nodes dying
or new nodes being deployed, signaling mechanisms
are needed in fixed assignment protocols to rectify
the assignment of resources to nodes.
• Typical protocols of this class are TDMA, FDMA,
CDMA, and SDMA.
9
2. Demand Assignment Protocols
• The allocation of resources to nodes is made on a short-
term basis, typically the duration of a data burst.
• This class of protocols can be further divided into
centralized and distributed protocols.
• In central control protocols, the nodes send out
requests for bandwidth allocation to a central node that
either accepts or rejects the requests.
• In case of successful allocation, a confirmation is
transmitted back to the requesting node along with a
description of the allocated resource.

10
3. Random Access Protocols
• Random access protocols incorporate a random element
by exploiting random packet arrival times, setting timers
to random values and so on.
• Typical random access protocols are pure ALOHA or
slotted ALOHA protocol, developed at the University of
Hawaii.
• In pure ALOHA protocol, a node willing for transmission
transmit a new packet it immediately.
• There is no coordination with other nodes and the
protocol thus accepts the risk of collisions at the receiver.
• To detect this, the receiver is required to send an
immediate acknowledgment for a properly received
packet.
• If no acknowledgement, the transmitter backs off for a
random time and starts the next trial.
11
 Topic 2

MAC Protocols for WSN

12
 Introduction
• The specific requirements and design considerations for
MAC protocols in wireless sensor networks are
explained below.
1. Balance of Requirements
2. Energy Problems on MAC Layer
– Collisions
– Overhearing
– Protocol Overhead
– Idle Listening

13
 Balance of Requirements
• The typical performance figures of WSN are –
1. Energy Efficiency
– New parameter
2. Fairness
– Not important because individual nodes do not
compete for bandwidth
3. Transmission delay
– Traded against energy conservation
4. Scalability & Robustness
– Important against changes in network topology

14
 Energy Problems on MAC Layer
• A node transceiver consumes a significant share of
energy.
• Moreover a transceiver can be in one of the four main
states - transmitting, receiving, idling, or sleeping.
• The important features are –
– Transmitting & receive costs similar
– Idling cheaper but as expensive as receiving
– Sleeping costs almost nothing but results in a “deaf”
node.
• The above features are applied to the operations of a
MAC protocol and the following energy problems and
design goals are derived.
15
1. Collisions
• Collisions incur useless receive costs at the destination
node, useless transmit costs at the source node, and
expend further energy upon packet retransmission.
Hence, collisions should be avoided, either by design or
by appropriate collision avoidance or hidden-terminal
procedures in CSMA protocols.
2. Overhearing
• Unicast frames have one source and one destination
node. But all the source’s neighbors in receive state can
hear a packet even though not destined to them. Hence
these nodes overhear the packet. For higher node
densities overhearing avoidance can save significant
amounts of energy.
16
3. Protocol overhead
• Protocol overhead is induced by MAC-related control
frames like RTS and CTS packets or request packets in
demand assignment protocols and by per-packet
overhead like packet headers and trailers.
4. Idle listening
• A node being in idle state is ready to receive a packet
but not currently receiving anything. This readiness is
costly and useless in case of low network loads.
Switching off the transceiver is a solution but mode
changes also cost energy, and their frequency should be
kept at “reasonable” levels. In case of TDMA-based
protocols, a node can exchange data only during
assigned time slot and switch off its transceiver in all
17
other time slots.
• In order to reduce the energy consumption with MAC
protocols in WSN –
– Expensive operations like complex scheduling
algorithms should be avoided.
– The desire to use cheap node hardware includes
components like oscillators and clocks.
– Frequent resynchronization of neighboring nodes
which can consume significant energy.

18
 Topic 3

Low Duty Cycle Protocols &

Wakeup Concepts

19
 Introduction
• Low duty cycle protocols avoid spending much time in
the idle state and reduce the communication activities
of a sensor node to a minimum.
• In an ideal case, the sleep state is left only when a node
is about to transmit or receive packets. A concept for
achieving this is the wakeup radio.

20
 Periodic Wakeup Scheme – Method 1
• First method is the cycled receiver approach illustrated
in Figure 3.2. In this approach, nodes spend most of
their time in the sleep mode and wake up periodically
to receive packets from other nodes.
• A node ‘A’ listens onto the channel during its listen
period and goes back into sleep mode when no other
node takes the opportunity to transmit a packet to A.
• A potential transmitter B must acquire knowledge
about A’s listen periods to send its packet at the right
time – this task corresponds to a rendezvous.
• This can be accomplished by letting node A transmit a
short beacon at the beginning of listen period to
indicate its willingness to receive packets.
21
 Periodic Wakeup Scheme - Method 2
• Second method is to let node B send frequent request
packets until one of them hits A’s listen period and is
answered by A.
• However, in both methods, node A only receives
packets during its listen period.
• If node A wants to transmit packets, it must acquire the
target’s listen period.
• The whole cycle consisting of sleep period and listen
period is also called a wakeup period.
• The ratio of the listen period length to the wakeup
period length is also called the node’s duty cycle.

22
 Important Observations from 2 Methods
• By choosing a small duty cycle, the transceiver is in
sleep mode most of the time, avoiding idle listening and
conserving energy.
• By choosing a small duty cycle, the traffic directed from
neighboring nodes to a given node concentrates on a
small listen period and in heavy load situations
significant competition can occur.
• Choosing a long sleep period leads to significant per-
hop latency.
• Sleep phases should not be too short, otherwise the
start-up costs outweigh the benefits.

23
 Comparison with other Protocols
• In other protocols like S-MAC, there is also a periodic
wakeup but nodes can both transmit and receive during
their wakeup phases.
• When nodes have their wakeup phases at the same
time, there is no necessity for a node wanting to
transmit a packet to be awake outside these phases to
rendezvous its receiver.

24
 Topic 4

S-MAC Protocol

25
 Introduction
• The S-MAC (Sensor-MAC) protocol provides mechanisms to
circumvent idle listening, collisions, and overhearing.
• S-MAC adopts a periodic wakeup scheme - each node
alternates between a fixed-length listen period and a fixed-
length sleep period as shown in Figure 3.3.
• However, the listen period of S-MAC can be used to receive
and transmit packets.
• S-MAC attempts to coordinate the schedules of
neighboring nodes such that their listen periods start at
the same time. A node x’s listen period is subdivided into
three different phases:
– Wakeup period
– Listen period
– Sleep period 26
Fig 3 / Principle of SMAC

27
 First Phase - Synch Phase
• During this phase, node x accepts SYNCH packets from its
neighbors.
• In these packets, the neighbors describe their own
schedule and x stores their schedule in a schedule table.
• Node x’s SYNCH phase is subdivided into time slots
according to a CSMA scheme.
• That is each neighbor y wishing to transmit a SYNCH
packet picks one of the time slots randomly and starts to
transmit if no signal was received in any of the previous
slots.
• In the other case, y goes back into sleep mode and waits
for x’s next wakeup.
• In the other direction, since x knows a neighbor y’s
schedule, x can wake at appropriate times and send its
own SYNCH packet to y.
28
 Second Phase - RTS Phase
• In the second phase (RTS phase), x listens for RTS
packets from neighboring nodes.
• In S-MAC, the RTS/CTS handshake is used to reduce
collisions of data packets due to hidden-terminal
situations.
• Again, interested neighbors contend in this phase
according to a CSMA scheme.

29
 Third Phase – CTS Phase
• In the third phase (CTS phase), node x transmits a CTS
packet if an RTS packet was received in the previous
phase. After this, the packet exchange continues,
extending into x’s nominal sleep time.
• In general, when competing for the medium, the nodes
use the RTS/CTS handshake whereby a node maintains
a NAV variable.
• The NAV mechanism can be used to switch off the node
during ongoing transmissions to avoid overhearing.

30
 Virtual Cluster Approach
• The S-MAC protocol allows neighboring nodes to agree
on the same schedule and to create virtual clusters.
• The clustering structure refers to the exchange of
schedules and the transfer of data packets is not
influenced by virtual clustering.
• The S-MAC protocol proceeds as follows to form the
virtual clusters -

31
1. A node x listens for a time at least the synchronization
period.
2. If x receives any SYNCH packet from a neighbor, it
adopts the announced schedule and broadcasts in one
of the neighbors’ next listen periods.
3. In the other case, node x picks a schedule and
broadcasts it.
4. If x receives another node’s schedule during the
broadcast packet’s contention period, it drops its own
schedule and follows the other one.
5. If node x already knows about the existence of
neighbors who adopted its own schedule, it keeps its
schedule.
32
 Message Passing Approach
• S-MAC also adopts a message-passing approach as
shown in Figure 3.4. In wireless media, it is advisable to
break a longer packet into several shorter ones. S-MAC
includes a fragmentation scheme working as follows.
– A series of fragments is transmitted with only one
RTS/CTS exchange between the transmitting node A
and receiving node B.
– After each fragment, B has to answer with an
acknowledgment packet. All the packets have a
duration field and a neighboring node C is required to
set its NAV field accordingly.

33
Fig 4 / SMAC Fragmentation & NAV Setting

34
• In S-MAC, the duration field of all packets carries the
remaining length of the whole transaction, including all
fragments and their acknowledgments.
• Therefore, the whole message shall be passed at once.
• If one fragment needs to be retransmitted, the
remaining duration is incremented by the length of a
data plus acknowledgement packet.

35
 Topic 5

The Mediation Device Protocol

36
 Introduction
• The mediation device protocol is compatible with the
peer-to-peer communication mode of the IEEE 802.15.4
WPAN standard.
• It allows each node in a WSN to go into sleep mode
periodically and to wake up only for short times to receive
packets from neighbor nodes.
• Each node has its own sleeping schedule and not take
care of its neighbors sleep schedules.
• Upon each periodic wakeup, a node transmits a short
query beacon indicating its willingness to accept packets
from other nodes.
• The node stays awake for some short time to open up a
window for incoming packets. If no packet is received
during this window, the node goes back into sleep mode.
37
 Mediation Device
• When a node wants to transmit a packet to a neighbor,
it has to synchronize with it.
• The dynamic synchronization approach achieves this
synchronization without requiring the transmitter to be
awake permanently to detect the destinations query
beacon.
• To achieve this, a mediation device (MD) is used.
• The mediation device is not energy constrained and can
be active all the time as shown in Figure 3.5.
• Because of its full duty cycle, the mediation device can
receive the query beacons from all nodes and learn
their wakeup periods.
38
The Mediation Device Protocol

39
 Dynamic Synchronization Approach
• Suppose node A wants to transmit a packet to node B.
The dynamic synchronization approach is given below-
– Node A announces this to the mediation device by
sending request to send (RTS) packets, which the MD
captures.
– There is a short answer window after the RTS packets,
where A listens for answers.
– After MD has received A’s RTS packet, it waits for B’s
next query beacon.
– The MD answers with a query response packet,
indicating A’s address and a timing offset, which lets B
know when to send the answering clear to send (CTS)
to A.
– Therefore, B has learned A’s period.
40
– After A has received the CTS packet, it can send its
data packet and wait for B’s acknowledgment.
– After transaction has finished, A restores its periodic
wakeup cycle and starts to emit query beacons
again.
– Node B also restores its own periodic cycle and thus
decouples from A’s period.

41
 Advantages
• It does not require any time synchronization between
nodes.
• The protocol is asymmetric that most of the energy
burden is shifted to the mediation device.
• The other nodes can be in the sleep state most of the
time and spend energy only for the periodic beacons.

42
 Disadvantages
• The nodes transmit their query beacons without
checking for ongoing transmissions.
• The beacons of different nodes may collide when nodes
have the same period and their wakeup periods overlap.

43
 Topic 6

Wakeup Radio Concepts

44
 Introduction
• The ideal situation will be if a node is always in the
receiving state when a packet is transmitted to it, in the
transmitting state when it transmits a packet, and in the
sleep state at all other times, the idle state should be
avoided.
• The requirement can be achieved by the concept of
wakeup radio by a simple, “powerless” receiver that
can trigger a main receiver if necessary.

45
 Proposed Wakeup Protocol
• The proposed wakeup MAC protocol comprises several
parallel data channels separated either in frequency
(FDMA) or in CDMA schemes.
• A node wishing to transmit a data packet randomly
picks one of the channels and performs a carrier
sensing operation.
• If the channel is busy, the node picks another random
channel and repeats the carrier-sensing operation.
• After a certain number of unsuccessful trials, the node
waits for a random time and starts again.
• If the channel is idle, the node sends a wakeup signal to
the intended receiver, indicating both the receiver
identification and the channel to use.
46
• The receiver wakes up its data transceiver, tunes to the
indicated channel, and the data packet transmission can
proceed.
• Afterward, the receiver can switch its data transceiver
back into sleep mode.

47
 Advantages
• Only low-power wakeup transceiver has to be switched
on all the time while more energy consuming data
transceiver can be non-sleeping particular node
involved in data transmissions.
• This scheme is naturally traffic adaptive, the MAC
becomes more and more active as the traffic load
increases.

48
 Disadvantages
• There is no real hardware for such low power wakeup
transceiver.
• The range of the wakeup radio and the data radio
should be the same.
• If the range of the wakeup radio is smaller than that of
data radio, all neighbor nodes cannot be woken up.
• If the range of the wakeup radio is larger, there can be a
problem with local addressing schemes
• This scheme depends on the wakeup channel’s ability
to transport useful information like node addresses and
channel identifications

49
 Topic 7

B-MAC Protocol
(Berkeley MAC Protocol)

50
 Introduction
• B-MAC is a widely used WSN MAC protocol. It is a part of
TinyOS.
• It employs low-power listening to minimize power
consumption due to idle listening.
• Nodes have a sleep period, after which they wake up and
sense the medium for preambles.
• If there is a preamble, the nodes stay awake and receive
the data packet after the preamble.
• If a node wants to send a message, it first sends a
preamble for the sleep period for all nodes to detect it.
After the preamble, it sends the data packet.
• There are optional acknowledgments also.
• After the data packet exchange, the nodes go back to
sleep.
• The preamble doesn’t contain addressing information.
Figure 3.6 shows the example transmission using B-MAC.51
Fig 6 / Example for BMAC Communication

52
 Preamble Sampling Scheme
• The B-MAC preamble sampling scheme checks the
channel for adjusting the time interval equal to frame
preamble size.
• As an example, if the medium is checked every 100 ms,
the preamble of the packet must be available for
100 ms at the minimum, in order for the receiver to
detect the packet.
• Upper layers may change the preamble duration,
according to the application requirements.

53
 Advantages
• It does not use RTS, CTS, ACK, or any other control
frame by default, but they can be added.
• It is one of the few specialized MAC protocols whose
implementation was tested in hardware.
• No synchronization is required and the protocol
performance can be tuned by higher layers to meet the
needs of various applications.

54
 Disadvantages
• The preamble creates large overhead.
• One example presents 271 bytes of preamble to send
36 bytes of data.

55
 Topic 8

IEEE 802.15.4. Standard

56
 Introduction
• The standard covers the physical layer and the MAC layer
of a low-rate Wireless Personal Area Network (WPAN).
• The targeted applications for IEEE 802.15.4 are in the area
of wireless sensor networks, home automation, home
networking, connect devices to PC etc.
• Most of these applications require only low-to-medium
bitrates, moderate delays and minimum energy
consumption.
• The physical layer offers bitrates of 20 kbps (a single
channel in frequency range 868–868.6 MHz), 40 kbps (ten
channels in range between 905 and 928 MHz) and 250
kbps (16 channels in range between 2.4 and 2.485 GHz).
• There are a total of 27 channels available, but the MAC
protocol uses only one of these channels at a time.
• The MAC protocol combines both schedule-based as well
as contention-based schemes.
57
 Network Architecture
• The standard distinguishes on the MAC layer two types
of nodes:
– A Full Function Device (FFD) can operate in three
different roles - a PAN coordinator, a simple coordinator
and a device.
– A Reduced Function Device (RFD) can operate only as a
device.
• A device must be associated to a coordinator node and
communicates only with this, forming a star network.
• Coordinators can operate in a peer-to-peer fashion and
multiple coordinators can form a Personal Area Network
(PAN).
• The PAN is identified by a 16-bit PAN Identifier and one
of its coordinators is designated as a PAN coordinator.
58
• A coordinator handles the following tasks:
– It manages a list of associated devices.
– It allocates short addresses to its devices
– In the beaconed mode, it transmits regularly frame
beacon packets announcing the PAN identifier, a list
of outstanding frames and other parameters.
– It exchanges data packets with devices and with peer
coordinators.

59
 Super Frame Structure
• The coordinator of a star network operating in the
beaconed mode organizes data transmission with the
help of a super-frame structure displayed in Figure 3.7.
• All super-frames have the same length. The coordinator
starts each super-frame by sending a frame beacon
packet. The various components of the following super-
frame are as follows-
– The super-frame is subdivided into an active period and
an inactive period. During the inactive period, all nodes
including the coordinator can switch off their
transceivers and go into sleep state.
– The active period is subdivided into 16 time slots.
60
Fig 7 / Super Frame Structure of IEEE 802.15.4

61
– The first time slot is occupied by the beacon frame and
the remaining time slots are partitioned into a
Contention Access Period (CAP) followed by a number
of contiguous Guaranteed Time Slots (GTSs).
– The length of the active period, inactive period, length
of a single time slot and the usage of GTS slots are
configurable.
– The coordinator is active during the entire active
period.
– The associated devices are active in the GTS phase only
in their allocated time slots.

62
 GTS Management
• The coordinator allocates GTS to devices only when the
latter send appropriate request packets during the CAP.
• One flag in the request indicates whether requested
time slot is a transmit slot or a receive slot.
• In a transmit slot, the device transmits packets to the
coordinator and in a receive slot the data flows in the
reverse direction.
• The coordinator answers the request packet in two
steps:
• An immediate acknowledgment packet with no
information about success or failure of the request.
• If resources available, the coordinator inserts a GTS
descriptor into one of the next beacon frames.
63
 Data Transfer Procedures
• Case 1: Device transmits data packet to coordinator
– If the device has an allocated transmit GTS, it wakes
up and sends its packet immediately without running
any carrier-sense operations.
– However, the device can do the full transaction only if
allocated time slots are available.
– When the device does not have any allocated slots, it
sends its data packet during the CAP using a slotted
CSMA protocol.
– The coordinator sends an immediate
acknowledgment for the data packet

64
• Case 2: Coordinator sends data packer to Device.
– If the device has allocated receive GTS and when the
packet/acknowledgment fits into these, the
coordinator simply transmits the packet in the
allocated time slot.
– The device has to acknowledge the data packet.
– When the coordinator is not able to use a receive
GTS, the handshake between device and coordinator
will happen as shown in Figure 3.8.

65
Fig 8 / Handshaking Operation

66
 Non-Beaconed Mode
• The coordinator does not send beacon frames nor any
GTS mechanism. No time synchronization exists.
• All packets from devices are transmitted without using
time slots because of lack of time synchronization.
• Coordinators must be switched on constantly but
devices can follow their own sleep schedule.

67
 Topic 9

Address and Name Management

68
 Introduction
• Naming and addressing are two fundamental issues in
networking.
• The names are used to denote things (for example,
nodes, data) whereas addresses supply the information
needed to find these things, for example, with routing in
a multi-hop network.
• Sometimes addresses are used to denote things too –
an IP address contains information to both find a node
and to identify a node more precisely a network
interface within a node.

69
 Use of Addresses & Names in Networks
1. Unique node identifier:
• A persistent data item unique for every node. An
example of a UID might be a combination of a vendor
name, a product name etc assigned at the time of
manufacturing.
2. MAC address:
• To distinguish between one-hop neighbors of a node.
This is important in wireless sensor networks using
contention-based MAC protocols
3. Network address:
• It is used to find and denote a node over multiple hops
and hence network addresses are often connected to
routing.
70
4. Network identifiers:
• To distinguish geographically overlapping wireless
networks of the same type and working in the same
frequency band.
5. Resource identifiers:
• It is represented in user-understandable terms. For
example, upon reading the name www.xemacs.org, an
experienced user knows that (i) the thing the name
refers to is likely a web server and (ii) the user can find
information about a great text editor.

71
 Address Management Tasks
1. Address allocation: Assignment of an address to an
entity from an address pool.
2. Address de-allocation: If the addresses of the dying
nodes were not put back into the address pool for
reuse, the address pool will be exhausted and no
addresses can be allocated to new nodes. Address de-
allocation can be either graceful or abrupt.
– Graceful de-allocation: A node sends out control
packets to give up its address.
– Abrupt de-allocation: The node disappears and does
not send appropriate control packets, leaving the
responsibility to the network.
72
3. Address representation: A format for representing
addresses needs to be negotiated and implemented.
4. Conflict Detection/Resolution: Address conflicts can
occur in networks with distributed assignment of on-
demand addresses or in case of mergers of so-far
distinct networks.
5. Binding: If several addressing layers are used, a mapping
between the different layers has to be provided. For
example, in IP networks, an IP address has to be
mapped to a MAC address using the ARP protocol.

73
 Uniqueness of Addresses
• Globally unique: A globally unique address is supposed
to occur at most once all over the world. An example is
48-bit IEEE MAC addresses used in Ethernet and Token
Ring networks.
• Network wide unique: A network wide unique address is
supposed to be unique within a given network, but the
same address can be used in different networks.
• Locally unique: A locally unique address might occur
several times in the same network, but it should be
unique within a suitably defined neighborhood.

74
 Address Allocation & Assignment
• The address assignment can happen a priori during the
manufacturing process or on demand, by using an
address assignment protocol.
• Such an on-demand address assignment protocol can be
either centralized or distributed.
• In a centralized solution, there is one single
authority/node taking care of the address pool, whereas
in distributed solutions, there is no such exposed node.
• The distinction between strong and weak Duplicate
Address Detection (DAD) are as follows:

75
1. Strong DAD:
• If address x is already assigned to node A at time t0 and
subsequently assigned to node B at time t1, then this
duplicate assignment must be detected latest at time t1
+ T where T is some fixed time bound.
2. Weak DAD:
• Duplicate addresses are tolerated as long as they do not
distort ongoing sessions. For example, if two networks A
and B merge and one address x is assigned in both
networks, no action should be taken as long as all
packets from nodes of the former network A destined to
x reach the node in A with address x and not the node
with the same address in the other network.
76
 Topic 10

Assignment of MAC Addresses

77
 Introduction
• The assignment of globally unique MAC addresses is
undesirable in sensor networks with mostly small
packets.
• A priori assignment of network wide unique addresses is
feasible only if it can be done with reasonable effort.
• But the overhead required to represent addresses is not
much large as in globally unique addresses.
• For example, up to 16,384 nodes can be addressed with
14 bits and this number is much friendlier than 48 bits
used for globally unique IEEE addresses.

78
Distributed Assignment of Network Wide Addresses
• A node chooses its address without any prior
information to use a uniform distribution on the address
range since this has maximum entropy.
1. Random address assignment
• Suppose that we have k nodes and each of these nodes
picks uniformly and independently a random address
from 0 to 2m − 1. The probability that these nodes
choose a conflict-free assignment has to be computed.
– For k = 1 this probability is one.
– For k = 2, the second node picks with probability (n−1/ n)
an address different from that of first node
– For k = 3, the third node picks with probability
[(n−1).(n−2) / n2] an address different from the first two
and so on.
79
• Hence the probability P (n, k) of the conflict-free
assignment is given by –

• Therefore, this method of random assignment quickly

leads to address conflicts.
• To preserve network wide uniqueness, either a conflict-
resolution protocol or more clever assignment schemes
should be chosen.

80
2. Techniques to deal with address collisions:
a. Auto-configuration Technique:
• A node starts by randomly selecting a temporary address
and a proposed fixed address and sends the address request
control packet.
• The temporary address is allocated from a dedicated
address pool and the routing protocol will find a path to a
node having the same fixed address.
• For such a node, an address reply packet is generated and
sent toward the temporary address.
• Upon receiving this reply, the node knows that the selected
fixed address is allocated and tries another address.
• If no address reply is received within a certain time, the
node repeats the address request packet a number of times
to compensate for possibly lost address reply packets.
• If still no address reply is received after all trials, the node
accepts the chosen IP address. 81
b. Initiator Technique:
• The initiator keeps a table of all known address
assignments and picks an unused address.
• The initiator then disseminates the proposed new address
to all nodes in the network and collects the answers. All
nodes put the proposed address into a list of candidate
addresses.
• If a node finds the address in the candidate list, it answers
with a reject packet, otherwise it answers with an accept
packet.
• If all known nodes have answered with an accept packet,
the initiator assigns the address to the requester and
informs all other nodes in the network that the
assignment now is permanent.
• Otherwise, the initiator picks another address and tries
again.

82
 Distributed Assignment Of Locally Unique Addresses
• This protocol assigns locally unique MAC addresses to
nodes by which a node communicates only with
immediate neighbors.
• Hence fewer bits are needed for address representation
than for network wide or globally unique addresses.
• By using locally unique addresses, the same address can
be used several times in the overall network. This
opportunity is taken by not transmitting addresses
directly but by encoding them and transmitting the code
words.
• The mapping from addresses to codes is called the
codebook and must be known a priori to nodes.
83
 Topic 11

Routing Protocols

84
 Introduction
• Routing strategies are required for transferring data
between the sensor nodes and the base station.
• Routing in WSN is different than traditional IP network
routing because it exhibits a number of unique
characteristics to build a global addressing scheme for a
large number of sensor nodes.
• Different routing techniques are proposed for remote
sensor network and these conventions can be classified
as per different parameters.
• The classification of routing methods is shown in Figure
3.10.

85
Fig 10 / Classification of Routing Protocols

86
 Functioning Mode Based Routing Protocols
• The function of a wireless sensor network specifies its
application. Hence routing protocols can be categorized
according to the operation used to satisfy a WSN function
as follows:
• Proactive Protocols: These protocols are also called as
table-driven protocols. In Proactive, the data is
transmitted to a BS through the predefined route.
Examples: LEACH, PEGASIS.
• Reactive Protocol: In Reactive Protocol the route is
established on demand. The route is established
dynamically when needed. Examples: TEEN, AODV, DSR
• Hybrid protocols: All the routes are found initially and
then improved at the time of sending data. These
protocols possess the concepts of both reactive and
proactive. For example APTEEN.
87
 Participation Style Based Routing Protocols
• Some WSNs consist of homogeneous nodes, whereas
some consist of heterogeneous nodes and these nodes
participate differently in every network according to
remaining energy of nodes, cluster head etc. Based on this
concept the protocols are classified as:
• Direct Communication protocols: In this type of protocols
the information sensed by nodes is sent directly to Base
Station (BS). Example: SPIN
• Flat protocols: In this, the nodes search for the valid path
and then transmit it to Base station. Example: Rumor
routing protocol.
• Clustering Protocols: In this, the area is divided into
clusters and Cluster heads are assigned to each cluster. All
the nodes in the cluster send data to corresponding
cluster heads and then cluster head sends it to Base
station. Example: TEEN
88
 Network Based Routing Protocols
• Network-based routing protocols depend on the
strategy how the network is prearranged. Such
protocols fall under three categories:
• Data Centric protocols: These are query based and they
depend on the naming of the desired data. The BS sends
queries within a certain region to get information and
waits for a reply from the nodes. Nodes in a particular
region collect the specific data based upon the queries.
Example: SPIN.
• Hierarchical protocols: In this, the nodes with lower
energy are used to capture information and nodes with
higher energies are used to process, transfer it and it is
used to perform energy efficient routing. Example:
TEEN, APTEEN.
89
• Location Based: In these, the location of nodes must be
known to find an optimal path using flooding. To get the
information about location of a particular node GPS is
used. Example: GEAR.

90
 Topic 12

Energy Efficient Routing

91
 Introduction
• Energy efficiency of a network is a significant concern in
wireless sensor network.
• These days networks are becoming large, information
gathered is becoming larger, which all consume a great
amount of energy resulting in an early death of a node.
• Therefore, many energy efficient protocols are
developed to lessen the power used in data sampling
and collection to extend the lifetime of a network.
• The following are some of energy efficient routing
protocols:

92
 1. LEACH – Low Energy Adaptive Clustering Hierarchy
• In this type of hierarchical protocol, most of the nodes
communicate to cluster heads. It consists of two phases:
• Setup Phase: In this phase, the clusters are ordered and
then Cluster Head (CH) has been selected. The task of
CH is to cumulate, wrapping, and forward the
information to the base station (Sink).
• Study State Phase: In this phase, the data is
communicated to the base station (Sink). To minimize
the overhead, the duration of this phase has been
increased. Each node in the network, contacts with the
cluster head, and transfer the data to it. Then CH will
develop the schedule to transfer the data of each node
to base station.
93
 2. PEGASIS [Power-Efficient Gathering in Sensor Information
Systems]
• It is a “chain-basis protocol” and an upgrading of the
“LEACH”.
• In “PEGASIS” every node transfers only with a close
neighbor to direct and obtain information. It turns
communicating to the BS, thus decreasing the quantity
of energy consumed per round.
• A chain should be developed, which can be completed
by the sensor nodes along with using an algorithm.
• On the other hand, the BS can compute this chain and
transmit it to all the sensor nodes.
• To develop the chain, all nodes have universal
information of the system and a greedy algorithm is
engaged
94
 3. Threshold sensitive Energy Efficient sensor Network
protocol
• The TEEN is a hierarchical protocol designed for the
conditions like sudden changes in the sensed attributes
such as temperature.
• The reduction of the number of transmissions is the
purpose of a hard threshold done by allowing the nodes
to transmit only when the sensed attribute is in the
range of interest.
• TEEN is well applicable for time important problems and
quite efficient in terms of saving energy and response
time.
• It also allows the user to manage the power utilization
and accurateness to suit the application.
95
 4. Adaptive Threshold sensitive Energy Efficient Sensor Network
• The “APTEEN” is an expansion of “TEEN” and goals at
both taking episodic data gatherings and replying to
time-critical events.
• As soon as the BS formulates the clusters, the CH
transmits the features, the values of threshold and
schedule of transmission to all nodes.
• After that, CH performs information accumulation in
order to preserve power.
• The main advantage of “APTEEN” in contrast to “TEEN”,
is that nodes utilize a smaller amount of power.
• The primary disadvantages of APTEEN are the
complication which results in lengthier deferment times.
96
 5. Directed Diffusion
• Directed diffusion is data-centric routing protocol for
collecting and publishing the information in WSNs.
• It has been developed to address the requirement of
data flowing from the sink toward the sensors.
• Its main objective is extending the network life time by
realizing essential energy saving.
• In order to achieve this, it has to keep the interactions
among the nodes within a limited environment by
message exchange.
• A localized interaction that provides multipath delivery
is a unique feature of this protocol.

97
 6. Energy Efficient Sensor Routing
• EESR is a flat routing algorithm proposed to decrease
the power utilization, data latency and to give scalability
in the WSN. It consists of Gateway, Base Station,
Manager Nodes, and Sensor Nodes. Their duties are-
• Gateway Delivers messages from Manager Nodes to the
Base Station, which has extra specification than normal
sensor nodes.
• It sends and receives messages to/from Gateway.
Moreover, it sends queries and collects data to/from
sensor nodes.
• Manager Nodes and Sensor Nodes collect data from the
environment and send it to each other in 1-Hop distance
till the Base Station.
98
 Typical Example for Energy Efficient Routing
• Energy-efficient unicast routing is a simple technique.
• It assigns to each link a cost value that reflects the
energy consumption across this link.
• It picks any one algorithm that computes least-cost
paths in a graph. (For Example: Shortest path algorithm
to obtain routes with minimal total transmission power.)
• An example scenario for a communication between
nodes ‘A’ and ‘H’ including link energy costs and
available battery capacity per node is shown in Figure
3.11.
• Various routes for communication between nodes ‘A’
and ‘H’ also show energy costs per packet for each link
and available battery capacity for each node 99
Fig 11 / Typical Example for EE Routing

100
• In the above example, it is found that -
• The minimum energy route is A-B-E-H which requires 3
units of energy only. The minimum hop count route
would be A-D-H which requires 6 units of energy.
• The following are the important parameters of Energy
efficient routing.
– Minimize Energy per packet
– Maximize network lifetime
– Routing considering available battery energy
– Maximum Total Available Battery Capacity
– Minimum Battery Cost Routing (MBCR

101
 Topic 13

Geographic Routing

102
 Introduction
• For many applications, it is necessary to address physical
locations as “any node in a given region” or “the node
at/closest to a given point”.
• When the position of source, destination and the
positions of intermediate nodes, are known, this
information can be used to assist in the routing process.
• The destination node has to be specified either
geographically or as some form of mapping.
• A source node knows the geographic area of the
destination and makes an impression on the destination
is called geographic routing.
• The area of the nodes is accessible through different
techniques like GPS, radio signal and so on.
103
 Importance of Geographic Routing
• Geographic routing (GR) is one in which a node
forwards a packet to the neighbor closest to the
destination.
• GR is an attractive approach for routing in WSNs due to
its low overhead and localized interactions.
• In GR, nodes will interact with their one-hop neighbors
to exchange the location information and make localized
forwarding decisions.

104
 Aspects of Geographic Routing
– First aspect: sending data to arbitrary nodes in a
given region, referred to as geo-casting.
– Second aspect: is called position-based routing or
“Cartesian routing”.
• In wireless sensor networks, usually the geo-casting
aspect of geographic routing is more important.
• Since nodes are considered as interchangeable and
distinguished by external aspects a location service is
not necessary.
• Hence, this concentrates on the geo-casting aspect, with
position-based routing aspects treated.

105
 Basics of Position Based Routing
• Assume a node wants to send a data packet to a node at
known position and every node in the network knows
its own position and that of its neighbors.
• In a simple greedy forwarding approach, the packet is
forwarded to that neighbor closest to the destination,
minimizing the remaining distance for the packet to
travel.

106
 Types of Geographic Routing
• Geographic routing has two parts-
– Geographic Forwarding
– Face Routing.
1. Geographic Forwarding:
• Geographic forwarding is a greedy routing algorithm
based on geography.
• For a given node, all its one-hop neighbors closer to the
sink belong to the forwarding set (FS).
• The node forwards an incoming data packet to the
neighbor in the FS closest to the sink.
• GR is attractive because it only requires nodes to maintain
the location information of their one-hop neighbors.
• Also, routing decisions can be made locally and 107
Fig 12 / Geographic Forwarding

108
2. Face Routing:
• However, Geographic Routing does not always succeed
in the greedy phase.
• When the forwarding node, e.g., node x in Figure 3.12,
has no one hop neighbor closer to the sink than itself, it
cannot further forward the incoming packet.
• Thus, the packet is stuck in a local minimum, called a
void, where the FS is empty.
• In such a case, a complementary mechanism called face
routing or backtracking towards a beacon is used to
route around the void.
• Utilizing the area data, the destination region is selected
and the packet is sent to that selected region
109
• This scheme overcomes the issue of limited power in
WSN as in some schemes nodes enters into the sleep
mode when they are not in used.
• The energy saving depends upon the number of dozing
nodes in the network. Fig 13 shows face routing below.

110
 List of Geographic Routing Protocols
• There are different sorts of geographic routing protocols
in wireless sensor networks with each having a different
feature of its own. Some of them are given as follows:
– MECN - Minimum Energy Communication Network
– GPSR - Greedy Perimeter Stateless Routing Protocol
– SMECN- Small Minimum Energy Communication
Network
– GEAR - Geographic Energy Aware Routing
– GAF - Geographic Adaptive Fidelity
– LAR – Location Aided Routing
– GOAFR - The Greedy Other Adaptive Face Routing
– TBF - Trajectory Based Forwarding
– SPAN- Coordination of Power Saving with Routing
111
 Greedy Perimeter Stateless Routing Protocol
• GPSR forwards a packet using greedy forwarding with
the “most forward” rule.
• If a packet cannot make any more progress, the packet
is switched to another routing mode called as perimeter
routing. A perimeter is a set of nodes defining a face.
• The perimeter routing consists of sending the packet
around the face using the right-hand rule.
• To do so, the packet carries information where it
entered a given face.
• This node ‘v’, the connecting line between ‘v’ and the
destination are used to decide whether the packet
should leave and proceed to the next one.
112
• The packet can return to greedy forwarding if the
distance of the current node to the destination and
node v has been effectively reduced.
• Figure 3.14 illustrates how a packet will be routed from
node A to node Z.
• While at node A, the packet can be greedily forwarded
to node D.
• At node D, greedy forwarding fails (both B and C are
further away from Z than D itself), so the packet has to
be routed round the perimeter of the interior face
defined by BFGCD.
• That is, it is forwarded to B and from there to F. Here,
edges F, G intersects line DZ and routing can proceed to
the next face.
113
Fig 14 / GPSR Protocol

114
• The packet proceeds around the perimeter of the
exterior face via E and I to H, from there via K to J and
then to L and Z.
• Since this face-based procedure is based on properties
of the plane, it only applies to planar graphs.
• In general, wireless network graphs are not planar.
• The performance guarantees of combined greedy/face
routing.
• When combining face routing and greedy routing, face
routing is tasked with routing around obstacles or out of
dead ends while greedy routing tries to make quick
progress toward the destination

115
Model Question Bank

116
 PART A
1. What is MAC protocol?
2. What is geographic addressing?
3. State the fundamental tasks of address management in
WSN.
4. Differentiate WSN routing with ad hoc routing.
5. What is energy efficient routing?
6. What is geographic routing?
7. Define Assignment of MAC address.
8. What is a Routing protocol?
9. Give the classes of MAC protocols.
10. What is Overhearing?
117
11. What is idle listening?
12. What is SMAC protocol?
13. What is a mediation device?
14. Mention the advantages of Mediation device protocol.
15. Give the demerits of Wake-up protocol.
16. Give the advantages of BMAC protocol.
17. What is IEEE 802.15.4 standard?
18. What is GTS management?
19. What are the techniques used for address collisions?
20. What is an Initiator?
21. What is LEACH?
22. Give the aspects of geographic routing.
23. Manton any four geographic routing.
24. What is GPSR protocol?
118
 PART B
1. Describe the MAC protocols for WSN in detail.
2. Write short notes on (i) SMAC and (ii) BMAC protocols.
3. Explain the IEEE 802.15.4 standard used for Wireless
Personal Area Network and its correlation with Zigbee.
4. Describe briefly the address and name management in
WSN.
5. Explain in detail about Energy efficient routing in WSN.
6. Explain in detail about Geographic routing in WSN.

119
 Wireless Sensor Networks

Unit 4 / Infrastructure Establishment

Prepared By

Dr. S.Omkumar

1
 Syllabus / Unit 4
• INFRASTRUCTURE ESTABLISHMENT:
• Topology Control, Clustering, Time
Synchronization, Localization and Positioning,
Sensor Tasking and Control.

2
 Topic 1

Topology Control

3
 Introduction
• In a dense wireless network, a single node has many
neighboring nodes with which direct communication
possible when using large transmission power.
• In turn, high transmission power requires lots of energy.
• Many neighbors are a burden for a MAC protocol and
routing protocols suffer from volatility in the network.
• To overcome these problems, topology control can be
applied.
• The idea is to restrict the set of neighbors of a given
node.
• This can be done by controlling transmission power,
introducing hierarchies in the network and turning off
some nodes for a certain time.
4
 Basic Ideas
• In a crowded network many wireless networking
problems are aggravated by the large number of
neighbors.
• Many nodes interfere with each other, a lot of possible
routes, nodes use large transmission power to talk to
distant nodes directly and routing protocols may re-
compute their routes.
• The above problems can be overcome by topology-
control techniques.
• The topology of a network is determined by the subset
of active nodes and the set of active links along which
direct communication can occur.
5
• A topology-control algorithm takes a graph G = (V, E)
representing the network where V is the set of all nodes
in the network.
• There is an edge (v1, v2) ∈ E if and only if nodes v1 and v2
can directly communicate with each other and
transforms it to a graph T = (VT ,ET).
• Figure 1 shows the topology for a dense WSN.

6
 Options for Topology Control
• To compute a modified graph T out of a graph G
representing the original network, a topology control
algorithm has the following options:
– The set of active nodes can be reduced by
periodically switching off nodes with low energy and
activating other nodes
– The set of active links or set of neighbors for a node
can be controlled.
– Active links/neighbors can also be rearranged in a
hierarchical network topology where some nodes
assume special roles.

7
 Aspects of Topology Control
• Connectivity: Topology control should not disconnect a
connected graph G
• Stretch factors: Removing links from a graph will
increase the length of a path between any two nodes u
and v. The hop stretch factor is defined as the worst
increase in path length for any pair of nodes u and v
between the original graph G and the topology-
controlled path T.
• Graph metrics: The importance of a small number of
edges in T and a low maximum degree (number of
neighbors) for each node.
• Throughput: The reduced network topology should be
able to sustain a comparable amount of traffic as the
original network.
8
• Robustness to mobility: When neighborhood
relationships change in the original graph G some other
nodes might have to change their topology information.
• Algorithm overhead: The overhead imposed by the
algorithm should be small.

9
 Controlling Topology – Power Control
• Controlling the set of neighbors to a particular node is the
basic approach of topology control.
• A flat topology is considered where all nodes are
operational and have the same tasks. This problem is
closely linked to controlling the transmission power of
nodes.
• When looking at the connectivity problem, there are two
options to approach the problem.
– Transmission range of a node
– Number of neighbors
• Under certain assumptions, these two options are
equivalent, but lead to different styles of proofs and
results.
• The first option, controlling the transmission range, is
purely geometric which leads to a uniform distribution of
nodes in a given area of size A.
10
 Some Example Protocols
1. Relative Neighborhood Graph (RNG)
• The Relative Neighborhood Graph (RNG) ‘T’ of a graph G
= (V,E) is defined as T = (V,E) where there is an edge
between nodes ‘u’ and ‘v’ if and only if there is no other
node ‘w’ closer to either ‘u’ or ‘v’ than ‘u’ and ‘v’ are
apart from each other.
• The RNG is easy to determine with a local algorithm.
• It is also necessarily connected if the original graph G is
connected.

11
Fig 2 / Construction of RNG

12
2. Spanning Tree–Based Construction
• Each node will collect information about its neighboring
nodes at maximum transmission power.
• A minimum spanning tree can be constructed for these
nodes, with energy costs used as link weights.
• The key is to maintain those edges in the reduced
topology correspond to direct neighbors in the minimal
spanning tree.
• This construction preserves the connectivity of the
original graph, and the maximum degree of each node.

13
Fig 3 / Spanning Tree Based Construction

14
3. A Distributed Common Power Protocol – COMPOW:
• The first observation is assigning the identical
transmission power to all nodes.
• The second observation is the need to keep the
transmission power level low for good connectivity.
• Each node determines routing tables for each
transmission power level.
• A node will use the smallest transmission power for
which the associated routing table has the same
number of entries as the table for the maximum
transmission power

15
Fig 4 / Greedy Algorithm for Minimum Power

16
 Topic 2

Clustering

17
 Introduction
• Clustering allows hierarchical structures to be built on
the nodes and enables more efficient use of resources
such as frequency spectrum, bandwidth, and power.
• For example, if the cluster size corresponds with the
direct communication range of the nodes then simpler
protocols can be used for routing and broadcasting
within a cluster.
• The same time or frequency division multiplexing can be
reused across non-overlapping clusters.

18
 Cluster Heads
• Clustering helps to monitor the health of the network
and misbehaving nodes.
• Networks can be comprised of mixtures of nodes having
special capabilities such as increased communication
range, GPS etc.
• These more capable nodes can play the role of cluster-
heads.
• However, the nodes are identical and their common
communication range is a natural cluster size.
• There exist several distributed protocols for cluster-head
election based on node unique identifiers (UIDs).

19
 Nominators
• It has a higher ID than all its “uncovered” neighbors—
neighbors already not claimed by another cluster-head.
• In others, each node nominates a cluster-head, the
highest ID node for communication.
• Nominated nodes then form clusters with their
nominators.

20
 Gateways
• Nodes that can communicate with two or more cluster-
heads may become gateways.
• They aid in passing traffic from one cluster to another
and also useful to view the IDs as weights for selecting
the cluster-heads.
• The number of clusters obtained is compared with the
minimum possible, even in a randomized setting.
• A constant approximation bound can be shown if the
leader election protocol is used hierarchically, with
increasing node ranges.

21
 Advantages of Clustering
• Clustering can be used to thin out parts of the network
where an excessive number of nodes present.
• Used to set up a simplified long range communication
network.
• Cluster-heads can have the minimum separation
comparable to the node communication range. This
property ensures that each cluster-head has a bounded
number of cluster-head neighbors.
• Several local communication protocols can become
simpler.

22
 Algorithm for determining Independent sets
• Consider a simple linear network shown as Figure 4.5.
• In step 1, nodes 2 and 5 cannot become cluster-heads
because their neighboring nodes 3 and 6 have not yet
decided and will take precedence over them.
• Once nodes 3 and 6 have learned about node 7 being a
cluster-head, they decide to become cluster members
and propagate this information to nodes 2 and 5.
• Then, these nodes can become cluster-heads in step 3.
• This essential algorithm has to be modified with the
following variations for useful in mobile networks.

23
– Whether to hold back nodes from forming clusters as
long as the cluster-head decision is revised, or to
allow intermediate clusters to be formed.
– Which nodes will re-clustered later
– Which may join another cluster-head

24
Fig 5 / Algorithm for determining
Independent Sets

25
 Topic 3

Time Synchronization

26
 Introduction
• Time synchronization allows for successful
communication between nodes on the network.
• Synchronization in wireless nodes implements a TDMA
algorithm over a multi-hop wireless network.
• Wireless time synchronization is used for many different
purposes including location, proximity, energy efficiency
and mobility.
• Time synchronization is used to save energy.
• It will allow the nodes to sleep for a given time and then
awaken periodically to receive a beacon signal.
• Many wireless nodes are battery powered, so energy
efficient protocols are necessary.
27
 Wireless Network Synchronization
• There are three basic types of synchronization methods
for wireless networks.
– The first is relative timing and the simplest which
depends on the ordering of messages and events.
– The next method is relative timing in which the
network clocks are independent of each other and
the nodes keep track of drift and offset.
– The last method is global synchronization where
there is a constant global timescale throughout the
network.

28
Fig 6 / Breakdown of Packet Delay Components

29
• As shown in Figure 6, all the wireless synchronization
schemes have four basic packet delay components:
• send time, access time, propagation time, and receive
time.
– The send time is the sender constructing the time
message to transmit on the network.
– The access time is the MAC layer delay in accessing
the network.
– The receive time is the receiving node processing the
message and transferring it to the host.

30
• There are many synchronization protocols which do not
differ much from each other.
– Reference Broadcast Synchronization (RBS),
– Timing-sync Protocol for Sensor Networks (TPSN)
– Flooding Time Synchronization Protocol (FTSP).
– Lightweight Time Synchronization Protocol (LTS)
• These protocols are the major timing protocols currently
in use for wireless networks.
• These protocols cover sender to receiver
synchronization as well as receiver to receiver.
• Also, they cover single hop and multi hop
synchronization schemes.

31
 Reference Broadcast Synchronization (RBS)
• RBS uses receiver to receiver synchronization and a third
party will broadcast a beacon to all the receivers.
• The beacon does not contain any timing information
and instead the receivers will compare their clocks to
one another to calculate their relative phase offsets.
• The timing is based on when the node receives the
reference beacon.
• The simplest form of RBS is one broadcast beacon and
two receivers.
• The timing packet will be broadcasted to the two
receivers. The receivers will record when the packet was
received according to their local clocks.
32
Fig 7 / Traditional Synch with RBS

33
• Then, the two receivers will exchange their timing
information and be able to calculate the offset.
• RBS can be expanded from the simplest form of one
broadcast and two receivers to synchronization between
‘n’ receivers; where n is greater than two.
• The main advantage of RBS is it eliminates the
uncertainty of the sender by removing the sender from
the critical path.
• By removing the sender, the only uncertainties are the
propagation and receive time.

34
 Time Synch Protocol for Sensor Networks
• TPSN is a traditional sender-receiver based synchronization
that uses a tree to organize the network topology.
• The concept is broken up into two phases, the level
discovery phase and the synchronization phase.
1. Level Discovery Phase:
• First, the root node should be assigned. If one node is
equipped with a GPS receiver, then it can be the root node
• Now the root node will send out the level discovery packet
to its neighboring nodes. The identity and level of the
sending node will be included in the packet.
• The neighbors of the root node will then assign
themselves as level one. They will in turn send out
the level discovery packet to their neighboring nodes.
35
• This process continues until all nodes received the level
discovery packet and are assigned a level
2. Synchronization Phase:
• Similar to the level discovery phase, the synchronization
phase begins at the root node and propagates through
the network.
• Figure 4.8 illustrates the two-way messaging between a
pair of nodes by following this method.
• The times T1, T2, T3, and T4 are all measured times.

36
Fig 8 / Two Communication between Nodes

37
 Flooding Time Synch Protocol (FTSP)
• It is similar to TPSN that it has a structure with a root node
and that all nodes are synchronized to the root.
• The root node will transmit the time synchronization
information with a single radio message to all participating
receivers.
• The message contains the sender's time stamp at
transmission. The receiver notes its local time when the
message is received.
• Having the sender's transmission time and the reception
time, the receiver can estimate the clock offset.
• FTSP was designed for large multi-hop networks. The root
is elected dynamically and responsible for keeping the
global time of the network.
• The network structure is mesh type topology instead of a
tree topology as in TPSN.
38
Fig 9 / Packets transmitted using FTSP

39
 Light Weight Time Synch Protocol
• The lightweight time synchronization (LTS) protocol is
meant to synchronize the clocks of a sensor network to
the clocks held by certain reference nodes having GPS
receivers.
• The protocol has control knobs to trade off energy
expenditure and achievable accuracy.
• LTS makes no restrictions with respect to the local clock
and not try to estimate actual drift rates.
• LTS subdivides time synchronization into two building
blocks:
– A pair-wise synchronization protocol synchronizes two
neighboring nodes.
– To keep all nodes synchronized to a common reference,
a spanning tree from the reference node to all nodes is
constructed.
40
 Topic 4

Localization & Positioning

41
 Introduction
• Localization is an important aspect in the field of
wireless sensor networks.
• The task of determining physical coordinates of sensor
nodes in WSN is known as localization or positioning
and is a key factor in today’s communication systems to
estimate the place of origin of events.
• Different localization methods are used in different
applications and there are several challenges in some
special scenarios such as forest fire detection.
• There are three approaches available to determine the
position of a node.
• They are –
42
1. Proximity:
Using information about the neighborhood of a node.
2. Triangulation and Trilateration:
Exploiting geometric properties of a given scenario
3. Scene Analysis:
To analyze characteristic properties of the position of a
node in comparison with premeasured

43
 Proximity
• It can be used to decide whether a node wants to
determine its position or location in the proximity of an
anchor.
• Typical example is the natural restriction of infrared
communication by walls, which can be used to provide a
node with simple location information about the room.
• Proximity-based systems are quite sophisticated and
used for approximate positioning when a node can
analyze proximity information of several overlapping
anchors .

44
 Trilateration (Determining the distances)
• The geometrical information can be used to derive
information about node positions.
• When distances between entities are used, the
approach is called Lateration and when angles between
nodes are used, it is called Angulation.
• In order to achieve perfections, the distance
measurements from more than three anchors can be
used, resulting in an approach called multilateration.
• To use multi-lateration, estimates of distances to anchor
nodes are required.
• This ranging process uses the facilities already present
on a wireless node, in particular, the radio
communication device.
45
• The characteristics of wireless communication are
determined by the distance between sender and
receiver at the receiver, they can serve as an estimator
of distance. The most important characteristics are –
1. Received Signal Strength Indicator (RSSI)
2. Time of Arrival (ToA)
3. Time Difference of Arrival (TDoA)

46
Fig 10 / Triangulation by intersecting 3 circles

47
1. Received Signal Strength Indicator (RSSI)
• If the transmission power PT, path loss model, and the
path loss coefficient ‘α’ are known, the receiver can use
the received signal strength ‘P’ to solve for the distance
‘d’ in a path loss equation given by –

• The distance estimates can be derived without

additional overhead from communication.
• The disadvantage is that RSSI values are not constant
but heavily oscillate, even when sender and receiver are
fixed.
• This is caused by effects like fast fading and mobility of
the environment
48
2. Time of Arrival (Time of Flight)
• If both sender and receiver know the time of
transmission, this parameter at the receiver can be used
to compute propagation time and distance.
• Depending on the transmission medium used, time of
arrival requires very high resolution clocks to produce
results of good accuracy.
• For sound waves, these requirements are modest and
very hard for radio wave propagation.

49
3. Time Difference of Arrival:
• The Time Difference of Arrival (TDoA) method directly
provides the start of transmission information to the
receiver.
• This can be done if two transmission mediums of very
different propagation speeds are used – for example,
radio waves propagating at the speed of light and
ultrasound, with a difference in speed.
• The disadvantage of this approach is the need for two
types of senders and receivers on each node.
• The advantage is better accuracy when compared to
RSSI-based approaches.

50
 Triangulation (Determining the Angles)
• Measuring angle can either be angle of a connecting line
between an anchor and a position-unaware node to a
given reference direction.
• It can also be angle between two such connecting lines
if no reference direction is commonly known to all
nodes as shown in Figure 4.11.
• A traditional approach to measuring angles is to use
directional antennas rotating on their axis, similar to a
radar station or a conventional lighthouse.
• This makes angle measurements simple, but
inappropriate for sensors nodes.

51
• In another technique, multiple antennas are mounted
on a device at known separation and measuring the
time difference between a signal’s arrival at the different
antennas, the direction from which a wave front arrived
at the device can be computed. Fig 11 shows angulation
between two anchors.

52
 Scene Analysis
• This method will analyze pictures taken by a camera and
derive the position from this picture.
• This requires computational effort and hardly
appropriate for sensor nodes.
• Apart from pictures, other measurable characteristic
“fingerprints” of a given location can be used for scene
analysis, for example, radio wave propagation patterns.
• One option is to use signal strength measurements of
one or more anchors transmitting known signal strength
and compare the actually measured values with those
that of stored in a database.

53
 Single Hop Localization
• Using the above basic building blocks of distance or
angle measurements a number of positioning or
locationing systems have been developed.
• In single-hop systems, a node with unknown position
can directly communicate with anchors.
• These single-hop systems predate wireless sensor
networks but provide the basic technology upon which
multi-hop systems are built.
• They are –

54
1. Active Badge
• The first system designed and built for locating simple,
portable devices – badges – within a building. Which
uses infrared as the transmission medium.
2. Active office
• This system targets the positioning of indoor devices.
Here ultrasound is used with receivers placed at well-
known position mounted in array at the ceiling of a
room.
3. RADAR
• This system is geared toward indoor computation of
position estimates. It employs the scene analysis
techniques, comparing the received signal
characteristics from multiple anchors with premeasured
and stored characteristic values.
55
4. Cricket
• Cricket is an example for systems which can compute their
own locations when privacy issues become relevant. It is
also based on anchors spread in a building, which provide
combined radio wave and ultrasound pulses to allow
measuring of signal strength information.
5. Overlapping Connectivity
• This is an example for an outdoor positioning system that
operates without any range measurements. It uses only
the observation of connectivity to a set of anchors to
determine a node’s position.
6. Approximate point in a Triangle
• The idea is to decide whether a node is within or outside
of a triangle formed by any three anchors. Using this
information, a node can intersect the triangles and
estimate its own position, similar to the intersection of
circles.
56
7. Using Angle of Arrival Information
• This method use anchors nodes that use narrow,
rotating beams where the rotation speed is constant
and known to all nodes. Nodes can measure the time of
arrival of each such beam, compute the differences
between two consecutive signals, and determine the
respective angles using geometric relationships

57
 Topic 5

Sensor Tasking & Control

58
 Introduction
• To efficiently utilize resources such as limited on-board
battery and limited bandwidth in a sensor network, sensor
nodes must carefully tasked and controlled to carry out
the required set of tasks.
• A utility-cost-based approach to sensor network
management is to address the balance between utility and
resource costs.
• The definitions for utility and cost are given below -
– Utility: Total utility of the data
– Cost: power supply and the communication bandwidth
• Among the total number of nodes, which sensor nodes are
to be activated and what information to transmit to the
network is a critical issue.
• This is because the sense values are not known and the
cost of sensing may vary with the data.
59
 Design Strategy for Sensor Tasking & Control
• The following are the various steps connected with
design strategy for sensor tasking and control -
– The important objects in the environment to be sensed
– The relevant parameters of these objects
– The relations among these objects critical to high level
information to be known
– The best sensor to acquire a particular parameter
– The sensing and communication operations needed to
accomplish the task
– The co-ordination given by the models of different
sensors
– The level of communicate information in a spectrum
from a signal to symbol
60
 Role of Sensor Nodes & Utilities
• A sensor may take on a particular role depending on the
application task requirement and resource availability
such as node power levels as shown in Figure 4.12. For
example -
– Nodes, denoted by SR, participate in both sensing
and routing.
– Nodes, denoted by S, perform sensing only and
transmit their data to other nodes.
– Nodes, denoted by R, decide to act only as routing
nodes, especially if their energy reserved is limited.
– Nodes, denoted by I, be in idle or sleep mode, to
preserve energy.
61
Fig 12 / Role of Sensor Nodes

62
 Information Based Sensor Tasking
• Information-based sensor tasking is to query sensors such
that information utility is maximized while minimizing
communication and resource usage.
• For localization or tracking problem, knowledge about the
target state such as position and velocity is required. This
requirement is represented as a probability distribution
over the state space in the probabilistic framework.
1. Sensor Selection:
• The estimation uncertainty can be approximated by a
Gaussian distribution, illustrated by uncertainty ellipsoids
in the state space.
• Sensor ‘b’ would provide better information than because
sensor ‘b’ lies close to the longer axis of the uncertainty
ellipsoid and its range constraint will intersect this longer
axis transversely. Figure 4.13 shows the sensor selection.63
Fig 13 / Sensor Selection based on Information Gain

64
• The following conditions are assumed. Figure 4.14
shows localizing a stationary source.
– All sensor nodes can communicate with each others.
– Sensor ‘a’ is farther from the leader node than the
sensor ‘b’
– There are four different criteria for choosing the next
sensor.
• Nearest Neighbor Data Diffusion
• Mahalanobis distance
• Maximum likelihood
• Best Feasible Region

65
Fig 14 / Localizing a Stationary Source

66
2. Algorithm for IDSQ
• A cluster leader selects optimal sensors to request data
from using the information utility measures.
• Using the Mahalanobis distance measure, the cluster
leader can determine which node can provide the most
useful information while balancing the energy cost,
without the need to have sensor data first.
• This algorithm is a single belief carrier node active at a
time.

67
 Joint Routing & Information Aggregation
• The primary purpose is to collect and aggregate
information.
• Information Driven Sensor Querying (IDSQ) just only
provides us with a method to obtain maximum
incremental information gain.
• There are some techniques to dynamically determine
the optimal routing path.
• The ellipses represent iso-contours of an information
field. The goal of routing is to maximally aggregate
information.
• This differs from routing in communication networks
where the destination is often known a priori to the
sender.
68
Fig 15 / Routing from a Query Proxy to high activity region

69
• The routing has to maximize information gain along the
path. A path toward the high information region may be
more preferable than the shortest path.
• Fig 16 Routing from a Query proxy to exit node.

70
Model Question Bank

71
 PART A
1. What is localization?
2. Give the advantages of localization.
3. What is topology?
4. Give the various aspects of topology.
5. Mention the various types of topology.
6. What is clustering?
7. List various services offered by localization.
8. Why is topology control necessary for WSN?
9. What are advantages of clustering?
10. What is positioning?
11. What is time synchronization?
72
12. List out the various synchronization protocols.
13. What are task driven in sensor nodes?
14. What is information based tasking?
15. What is sensor tasking?
16. What is sensor control?
17. What is synchronization?
18. What is FTSP protocol?
19. What is triangulation?
20. What is proximity?
21. Mention the important characteristics of tri-lateration.
22. What is called active badge?

73
 PART B
1. Describe the various aspects and options for Topology
control in WSN with relevant example protocols.
2. Explain the concept of Localization and Positioning in
detail.
3. Discuss in detail the various algorithms of Time
Synchronization.
4. Explain the concept of Clustering, its advantages and
algorithm for determining independent sets.
5. Write a short note on Sensor Tasking and Control.

74
 Wireless Sensor Networks

Unit 5 / Sensor Network Platforms &

Tools

Prepared By

Dr. S.Omkumar

1
 Syllabus / Unit 5
SENSOR NETWORK PLATFORMS & TOOLS:
• Sensor Node Hardware – Berkeley Motes,
Programming Challenges, Node-level software
platforms, Node level Simulators, State-centric
programming

2
 Topic 1

Sensor Node Hardware – Berkeley

Notes

3
 Introduction
• A real-world sensor network application has to
incorporate all elements subject to energy, bandwidth,
computation, storage, and real-time constraints.
• This makes sensor network application development
quite different from traditional distributed system
development or database programming.
• A sensor network application can assume an always-on
infrastructure that provides reliable services such as
optimal routing, global directories, or service discovery.
• There are two types of programming for sensor
networks, those carried out by end users and those
performed by application developers.
4
• An end user may view a sensor network as a pool of
data and interact with the network via queries.
• An application developer must provide end users of a
sensor network with the capabilities of data acquisition,
processing, and storage.

5
 Sensor Node Hardware
• A typical sensor node hardware contains a general-
purpose CPU and working memory, long-term stable
storage such as flash memory or disk and I/O
capabilities to support sensors.
• Sensor nodes have evolved into two broad categories:
– Small devices with 8-bit microcontrollers as CPUs,
10–100 KB of working memory and 100–1000 KB of
flash secondary storage
– Larger devices with 32-bit CPUs, megabytes each of
working memory and secondary storage

6
 Small Devices (Berkeley Notes)
• The Berkeley motes are a family of embedded sensor
nodes sharing roughly the same architecture. Let us
take the MICA mote as an example.
• The MICA motes have a two-CPU design. The main
microcontroller (MCU), an Atmel ATmega103L takes
care of regular processing.
• A separate and less capable coprocessor is only active
when the MCU is being reprogrammed.
• The ATmega103L MCU has integrated 512 KB flash
memory and 4 KB of data memory.
• In addition to that, a MICA mote also has a separate
512 KB flash memory unit for holding data.
7
Fig 1 / MICA Mote Architecture

8
• Since the connection between the MCU and this
external memory is via a low-speed serial peripheral
interface (SPI) protocol, the external memory is more
suited for storing data than for storing programs.
• The RF communication on MICA motes uses the TR1000
chip set operating at 916 MHz band.
• It can achieve a maximum of 50 kbps raw data rate.
MICA motes implement a 40 kbps transmission rate.
• The maximum transmission range is about 300 feet in
open space.
• MICA motes support a 51 pin I/O extension connector.
Sensors, actuators, serial I/O boards, or parallel I/O
boards can be connected via the connector
9
• A sensor/actuator board can host a temperature sensor,
a light sensor, an accelerometer etc.
• The serial I/O connection allows the mote to
communicate with a PC in real time.
• The parallel connection is primarily for downloading
programs to the mote.
• A radio transmission bears the maximum power
consumption.
• However, each radio packet (e.g., 30 bytes) only takes 4
ms to send, while listening to incoming packets turns
the radio receiver ON all the time.
• There are huge differences among the power
consumption levels in active mode, idle mode and
suspend mode of the MCU.
10
 Large Devices
• The larger class of devices is exemplified by products
such as Stargate designed by Intel and Cerfcube from
Intrinsyc.
• These devices are used in a variety of embedded
applications.
• In the sensor network, they are used as gateways to a
collection of motes, or for applications that require
heavier-duty signal processing.
• Each such device employs an X-Scale or ARM-based
processor which has 64MB working memory and 1GB
flash-based secondary storage.
• They support many connectivity options including USB,
IEEE 802.11 wireless and a 51-pin mote connector
allowing use of a mote and its radio.
11
 Topic 2

Programming Challenges

12
 Introduction
• A sensor network differs from traditional computing
environments in various aspects, thereby insisting
programming frameworks and tools required for a
sensor network.
• Specifically, the following characteristics affect the
design of sensor network programming tools:
– Reliability
– Resource Utilization
– Scalability
– Data Centric Networks

13
 Reliability
• Wireless sensor networks are more unreliable than
other distributed systems.
• Therefore, sensor networks are built to adapt to
changing dynamics and node & link errors such that
network continues to serve until the network fails.
• While many faults in a network will never be noticed by
application, resilience to failures and topology changes
should be supported by a programming environment.

14
 Resource Constraints
• Wireless sensor networks are very resource-constrained
which affects the programming approach, maximum
code size and other aspects of application
development.
• Energy efficiency is particularly critical in WSNs and
penetrates every aspect of sensor network design from
duty cycles to routing protocols.
• Therefore, programming tools and models should be
developed to exploit energy-saving techniques and
approaches.

15
 Scalability
• Sensor networks can scale up to hundreds and
thousands of sensor nodes.
• Therefore programming models should support
developers in designing applications and software for
large-scale and heterogeneous networks.
• Manual configuration, maintenance and repair of
individual sensor nodes will be infeasible due to the
large number of devices, therefore insisting support for
self-management and self-configuration.
• The scale of a network can also be addressed by using
programming models that consider entire network as a
single entity instead of focusing on each individual
device.
16
 Data Centric Networks
• In many wireless sensor networks, not only are the
individual sensor nodes of interest, but also the data they
generate and disseminate.
• Sensor network applications obtain useful information in
a timely fashion. Many applications are only concerned
with the collection of data at a central point.
• For example, a server that stores, analyzes, or visualizes
the sensor data.
• Other applications require immediate processing and
analysis of data within the network to eliminate
redundant information, to aggregate data from multiple
sensors etc.
• Each category will require different programming models
so that programming a network results in generating
distributed algorithms to work across many nodes in an
efficient manner.
17
 Topic 3

Node Level Software Platforms

18
 Introduction
• The following are the node level software platforms
used in wireless sensor networks –
– TinyOS
– NesC
– TinyGALS
– Sensor Network Application Construction Kit (Snack)
– Thread Based Model

19
1. TinyOS

20
 Introduction
• TinyOS is an embedded component-based operating
system and platform for low-power wireless devices
in wireless sensor networks (WSNs), ubiquitous
computing, personal area networks, building automation
and smart meters.
• It is written in the programming language nesC, as a set of
cooperating tasks and processes.
• It was collaboration between the University of California,
Berkeley, Intel Research and Crossbow Technology.
• It was released as open-source software under BSD license
and has grown into an international consortium,
the TinyOS Alliance.
• TinyOS has been used in space, being implemented
in ESTCube-1.
21
 Features of TinyOS
• TinyOS applications are written in the programming
language nesC a dialect of C language optimized for the
memory limits of sensor networks.
• Its supplementary tools are mainly in the form
of Java and shell script front-ends.
• The associated libraries and tools are mostly written in
C.
• TinyOS programs are built of software components and
they present hardware abstractions.
• Components are connected to each other
using interfaces.
• TinyOS provides interfaces and components for
common abstractions such as packet communication,
routing, sensing, actuation and storage.
22
 Characteristics of TinyOS
• TinyOS is fully non-blocking and it has one call stack.
• All I/O operations are asynchronous and have a callback.
• TinyOS uses nesC's features to link these callbacks called
events to enable the native compiler for better
optimization,
• TinyOS forces programmers to write complex logic by
combining together many small event handlers.
• TinyOS provides tasks similar to a Deferred Procedure
Call in order to support larger computations,.
• Tasks are non-preemptive and run in first in first
out order.
• This simple concurrency model is sufficient for I/O centric
applications but difficult with CPU-heavy applications.
23
Fig 2 / Field Monitor Example

24
 Application Example
• Let us consider a Tiny OS application example—Field
Monitor, where all nodes in a sensor field periodically
send their temperature and photo sensor readings to a
base station via an ad hoc routing mechanism.
• A diagram of the Field Monitor application is shown in
Figure 5.2, where blocks represent TinyOS components
and arrows represent function calls among them.
• The directions of the arrows are from callers.

25
2. NesC

26
 Introduction
• The name nesC is an abbreviation of "network
embedded systems C".
• NesC is a component-based event-driven programming
language used to build applications for TinyOS platform.
• TinyOS is an operating environment designed to run on
embedded devices used in distributed wireless sensor
networks.
• NesC is built as an extension to the C programming
language with components "wired" together to run
applications on TinyOS.

27
 Components
• NesC programs are built out of components assembled
to form whole programs.
• Components have internal concurrency in the form of
tasks.
• Threads of control may pass into a component through
its interfaces. These threads are rooted either in a task
or a hardware interrupt.
• Interfaces may be provided or used by components.
• The provided interfaces are intended to represent the
functionality that the component provides to its users.
• The used interfaces represent the functionality the
component needs to perform its job.
28
 Interfaces
• In nesC, interfaces are bidirectional.
• They specify a set of functions to be implemented by the
interface's provider (commands) and a set to be
implemented by the interface's user (events).
• This allows a single interface to represent a complex
interaction between components.
• This is critical because all lengthy commands in TinyOS are
non-blocking and their completion is signaled through an
event.
• Typically commands call downwards, i.e., from application
components to those closer to the hardware, while events
call upwards.
• Components are linked to each other via their interfaces.
• This increases runtime efficiency, encourages robust
design, and allows for better static analysis of programs.
29
 Timer Service
• In order to connect the Timer component and a
hardware clock wrapper called HWC lock, a timer
service called Timer C is provided as shown in Figure
5.3.
• The Timer interface defines two types of commands:
start and stop.
• The Timer interface also defines an event, which is also
a function.
• While commands are implemented by the providers of
an interface, events are implemented by the users.
• Similarly, all other interfaces in this example define both
commands and events.
30
Fig 3 / Timer C Configuration

31
 Components Implementation
• Modules are components implemented by application
code, while configurations are components
implemented by connecting interfaces of existing
components.
• Every nesC application has a top-level configuration that
describes how components are wired together.
• Functions in nesC are described as (f. i) where f is a
function in an interface i.
• Functions are invoked using the call operation (for
commands) and the signal operation (for events).

32
 Classification of Codes
• In TinyOS, code executes either asynchronously or
synchronously.
• Race conditions can occur when concurrent updates to
shared state are performed.
– Asynchronous code (AC): Code reachable from at least
one interrupt handler
– Synchronous code (SC) : Code only reachable from
tasks
• Synchronous code is always atomic to other synchronous
codes because tasks are always executed sequentially
without preemption.
• Race conditions are possible when shared state is
modified from AC or when shared state from SC is also
modified from AC. 33
• Therefore, nesC provides programmers with two
options to ensure atomicity.
• The first option is to convert all of the sharing code to
tasks.
• The second option is to use atomic sections to modify
shared state.

34
3. TinyGALS

35
 Introduction
• TinyGALS is a globally asynchronous and locally
synchronous (GALS) approach for programming event-
driven embedded systems.
• A TinyGALS program consists of modules, which are
composed of components.
• A component C has a set of internal variables VC, a set of
external variables XC and a set of methods IC that operate
on these variables.
• Methods are further divided into calls in the ACCEPTSC set
(which can be called by other components) and calls in the
USESC set (which are needed by C and may belong to other
components).
• Similar to nesC and TinyOS, TinyGALS defines components
using an interface definition and an implementation.
36
 Composition of Modules
• TinyGALS modules consist of one or more components.
A module ‘M’ is a 6-tuple given by –
• M=(COMPONENTSM, INITM, INPORTSM, OUTPORTSM,
PARAMETERSM, LINKSM)
where -
– COMPONENTSM is the set of components of M
– INITM - list of methods of M’s components,
– INPORTSM - inputs of module
– OUTPORTSM - outputs of the module,
– PARAMETERSM - a set of variables external to the
components
– LINKSM - relationships between the method call
interfaces and I/O of the module.
37
• Modules are further connected to each other to form a
complete TinyGALS system with a 5-tuple ‘S ‘given by –
• S=(MODULESS, GLOBALSS, VAR_MAPSS, CONNECTIONSS,
STARTS)
where -
– MODULESS – a set of modules
– GLOBALS – a set of global variables
– VAR_MAPSS – a set of mappings
– CONNECTIONSS - a list of the connections between
module output ports and input ports
– STARTS - name of an input port of one module used
as a starting point for the execution of the system.
38
 TinyGUYS
• The architecture of TinyGALS can be used to automate
the generation of scheduling and event handling code.
• By this, error-prone concurrency control code can be
avoided.
• Code generation tools can automatically produce all
necessary code for component links, module
connections, system initialization etc.
• The use of message passing, modules in TinyGALS
become decoupled from each other, therefore
facilitating their independent development.
• TinyGALS provides another mechanism, called
TinyGUYS (Guarded Yet Synchronous) variables, where
modules may read global variables synchronously
without delay.
39
4. Sensor Network Application
Construction Kit (SNACK)

40
 SNACK
• The Sensor Network Application Construction Kit
(SNACK) is a configuration language component and
service library and compiler for the development of
sensor network applications.
• SNACK’s goal is to provide smart libraries that can be
combined to form sensor network applications.
• This will simplify the development process without
losing control over efficiency.
• The SNACK library of components and services contains
a variety of components for sensing, aggregation,
transmission, routing, and data processing

41
5. Thread Based Model

42
 Thread Based Model
• The thread-based paradigm is popular in many computing
systems and now sensor networks.
• In traditional event-based systems, event handlers are
executed in response to events and these handlers (tasks)
run to completion without interruption from other tasks.
• The main advantage of the thread-based approach is
multiple tasks can make progress in their execution
indefinitely.
• For example, a task scheduler can execute a task for a
certain amount of time; preempt this task in order to
execute another task.
• This time-slicing approach simplifies the programming of
sensor systems at the cost of increased operating system
complexity
43
 Topic 4

Node Level Simulators

44
 Introduction
• Node-level design methodologies are usually associated
with simulators that simulate the behavior of a sensor
network on a per-node basis.
• Using simulation, designers can quickly study the
performance in terms of timing, power, bandwidth, and
scalability of potential algorithms without implementing
them on actual hardware.
– NS2
– GloMoSim and QualNet
– JiST/SWANS
– OMNeT++
– TOSSIM
– EmStar
– Avrora
45
 NS2
• It was written in a combination of C++ and an object-
oriented dialect of Tcl called OTcl.
• Many enhancements and extensions were developed to
provide support for wireless networks and mobile adhoc
networks.
• Similarly, a variety of extensions for sensor networks have
been created. For example, one such extension adds the
concept of a phenomenon to a sensor network simulation.
• The model uses broadcast packets transmitted through a
designated channel to represent a phenomenon.
• Broadcast packets are generated using the PHENOM
routing protocol, which emits packets with a certain
configurable pulse rate
46
 GloMosim & QualNet
• GloMoSim is a simulation tool based on the PARSEC
simulation environment.
• Parallel Simulation Environment for Complex (PARSEC)
systems is a C-based simulation language used to
represent a set of objects in the physical system as
logical processes and interactions among these objects
as time-stamped message exchanges.
• GloMoSim supports a variety of models at different
protocol layers such as CSMA & MACAW (MAC layer),
flooding & DSR (network layer) and TCP & UDP
(transport layer).

47
• In addition, it supports different node mobility models
of the following -
• Random waypoint model: A node chooses a random
destination within simulated area and moves toward
this destination with a specified speed
• Random drunken model: A node periodically moves to
a position chosen randomly from its immediate
neighboring positions

48
 Jist/SWANS
• JiST stands for Java in Simulation Time.
• The motivation behind JiST is to create discrete event
simulations that execute efficiently and transparently.
• Efficiency refers to the ability to execute a given
simulation program in parallel, while optimizing the
configuration of the simulation across the available
computational resources.
• Transparency refers to the ability to transform simulation
programs automatically to run with simulation time
semantics.
• The primary motivation for JiST was to support
simulations of ad hoc networks and Scalable Wireless Ad
hoc Network Simulator (SWANS).
• SWANS is a collection of independent software
components aggregated to form complete wireless
simulations.
49
 OMNeT++
• The Objective Modular Network Testbed (OMNeT++)
discrete event simulation environment is a tool used for
the simulations of communication networks,
Multiprocessors and various distributed systems.
• It is an open-source simulator based on C++ designed
for the simulation of large systems and networks.
• A model in OMNeT++ consists of modules that
communicate with each other using message passing.
• Simple modules can be grouped together to form more
complex compound modules.
• A user defines the structure of a module using
OMNeT++’s topology description language NED.
• The OMNeT++ framework includes a graphical editor
used to edit network topologies either graphically or in
NED source view.
50
 TOSSIM
• A simulator for TinyOS-based wireless sensor networks is
TOSSIM.
• It generates discrete event simulations directly from
TinyOS components, running the same code on sensor
nodes.
• TOSSIM replaces low-level components for translating
hardware interrupts into events in the simulation that
drive the TinyOS application.
• TOSSIM works at the bit level. This allows for
experimentations with low-level protocols in addition to
higher-level protocols or applications.
• Similar to most other tools, TOSSIM comes with a
visualization tool called TinyViz.
• TOSSIM scales to thousands of sensor nodes and its
advantages include scalability and extensibility.
51
 EmStar
• EmStar is targeted at high capability nodes called
microservers.
• These nodes in a hierarchical sensor network structure
run more complex software than ordinary sensing
devices.
• EmStar consists of a Linux microkernel extension,
libraries, services, and several tools.
• EmSim operates many virtual nodes in parallel in a
simulation that models radio and sensor channels.
• EmCee runs the EmSim core and is an interface to real
low-power radios instead of using a modeled channel.
• EmView is a graphical visualizer for EmStar systems.
52
 Avrora
• Avrora is a flexible simulator framework implemented
in Java.
• Each node is implemented as its own thread and code is
executed in an instruction-by instruction fashion.
• The key component is its implementation of an event
queue.
• Many energy-conscious nodes tend to sleep for large
periods of time where no instructions are executed and
the energy consumption is reduced.
• The event queue in Avrora takes this advantage to
boost the performance of the simulator.

53
 Topic 5

State Centric Programming

54
 Introduction
• Applications such as target tracking are not generic
distributed programs.
• The vital parameters are the states of physical
phenomena and models of their evolution over space
and time.
• A distinct property of physical states such as location,
shape and motion of objects is continuity in space and
time.
• The sensing and control of these states can be handled
through sequential state updates.

55
• The System theories provide the following state centric
abstraction for state updating:
– xk+1 = f(xk, uk) (1)
– yk = g(xk, uk) (2)
• where ‘x’ is the system state, ‘k’ is an integer update index
over time, ‘u’ is input, ‘y’ is output, ‘f’ is the state update
function and ‘g’ is the output or observation function
• This formalization is broad enough to capture a wide
variety of algorithms in sensor fusion, signal processing
and control.
• State-centric programming abstractions have been
successfully applied to synchronous VLSI circuit designs
and control system designs.
56
 Domain Specific Run Time Systems
• Synchronous languages such as Signal & Esterel and
mixed-signal visual languages such as Matlab’s Simulink &
Ptolemy are all examples of state-centric programming
models.
• However in a distributed real-time embedded system the
formulation is not cleanly represented.
• The relationship among subsystems can be highly
dynamic. The following concerns must be addressed –
– The storage of state variables
– Origin of inputs
– Destination of outputs
– Evaluation of functions ‘g’ and ‘f’
– Time taken to acquire the set of inputs
– The correct order of arrival of inputs
– The choice of update interval
57
• System designers cannot be shielded from these issues
without compromising system correctness and
efficiency.
• These concerns must be addressed to perform sensing,
computation, actuation and play a central role in
achieving the overall system performance.
• However, traditional programming models and
languages don’t support these “nonfunctional” aspects
of computation. The novel design methodologies and
frameworks are required to provide meaningful
abstractions for these issues.
• Domain-specific runtime systems are used to support
this design methodology to ensure efficient execution
and allow transparent features such as security and
reliable communication.
58
 Collaborative Groups
• A group is a set of entities that contribute to a state
update.
• These entities can be physical or logical sensor nodes or
more abstract system components such as software
agents.
• A group encapsulates two properties – its scope and
structure. A group’s scope defines its members.
• The scope can be specified by a membership function.
• Grouping nodes according to physical attributes rather
than node addresses is an important abstraction in
sensor network programming.
• The scope can be evaluated locally or dynamically, as
long as communication among the group members is
maintained.
59
 Examples for Groups
1. Geographically Constrained Group (GCG)
• GCG consists of members within a pre-specified
geographical extent.
• Since physical signals from point targets, may propagate
only to a limited extent in an environment, this kind of
group represents all the sensor nodes that can possibly
“sense” a phenomenon.
• There are many ways to specify the geographic shape,
such as circles, polygons, and their unions and
intersections.
• A GCG can be easily established by geographically
constrained flooding.
60
2. N-hop Neighborhood Group (n-HNG)
• When the communication topology is more important
than the geographical extent, hop counts are useful to
constrain group membership.
• An n-hop neighborhood group has an anchor node and
defines that all nodes within n communication hops are
members of the group.
• Since it uses hop counts rather than Euclidean
distances, local broadcasting can be used to determine
the scope.
• The anchor node is the leader of the group and the
group may have a tree structure with the leader as the
root to optimize for communication.
61
3. Publish/Subscribe Group (PSG)
• A publish/subscribe group comprises consumers
expressing interest in specific types of data or services
and producers that provide those data or services.
• Communication among members of a PSG may be
established via rendezvous points, directory servers, or
network protocols such as directed diffusion.

62
4. Acquaintance Group
• In this group, a member belongs to the group because
it was “invited” by another member in the group.
• The relationships among the members may not depend
on any physical properties at the current time.
• A member may also quit the group without requiring
permission from any other member.
• An AG may have a leader, serving as the rendezvous
point.
• When the leader is also fixed on a node, GPSR, ad hoc
routing trees may facilitate the communication
between the leader and the other members.
• The important use of this group is to monitor and
control mobile agents from a base station
63
5. Using Multiple types of Groups
• Mixing and matching groups is a powerful technique for
tackling system complexity by making algorithms much
more scalable and resource efficient without sacrificing
conceptual clarity.
• The highly tuned communication protocols can be used
for specific groups to reduce latency and energy costs.
• There are various ways to compose groups.
• They can be composed in parallel to provide different
types of input for a single computational entity.

64
 State Centric Design Framework (PIECES)
• Programming and Interaction Environment for
Collaborative Embedded Systems (PIECES) is a software
framework that implements the methodology of state-
centric programming over collaboration groups to
support the modeling, simulation and design of sensor
network applications.
• It is implemented in a mixed Java-Matlab environment.
PIECES comprise principals and port agents.
• Figure 5.4 shows the basic relations among principals
and port agents.

65
Fig 4 / Principals & Port Agents

66
• The features of a Principal are as follows-
– The key component for maintaining a piece of state.
– To update its state from time to time, a computation
corresponding to evaluating function ‘f’.
– Accepts other principals’ queries of certain views on
its own state, a computation corresponding to
evaluating function ‘g’.
– Gather information from other principals for
updating
– Creates port agents and attaches them onto it and
onto the other principals.

67
• The features of a port are as follows-
– May be an input, an output, or both.
– Also called an observer, computes outputs based on
the host principal’s state and sends them to other
agents.
– May be active or passive.
– Active observer pushes data autonomously to its
destination, while a passive observer sends data only
when a consumer requests it.
– Capture communication patterns among principals.
• The execution of principals and port agents can be
either time driven or event-driven where events may
include physical events that are pushed to them or
query events from other principals or agents.
68
 Multi-target Tracking System
• Tracking of two crossing targets can be decomposed
into three phases:
– When the targets are far apart, the tracking problem
can be treated as a set of single-target tracking sub-
problems.
– When the targets are in proximity of each other, they
are tracked jointly due to signal mixing.
– After the targets move apart, the tracking problem
becomes two single-target tracking sub-problems
again.

69
Model Question Bank

70
 PART A
1. What is sensor node hardware?
2. What is TinyOS?
3. Where is TinyOS used?
4. Classify the sensor node hardware.
5. Define Berkley notes.
6. What do you mean by node level simulation?
7. What are the programming challenges for sensor
networks?
8. What are the different platforms available for sensor
networks?
9. What is node level software platform?
10. What is a node level simulator?
71
11. What is centric programming?
12. List the major concern of sensor node hardware.
13. What is TINYGALS?
14. Highlight the salient feature of component-based
operating system.
15. What is PIECES?
16. Give the features of a principal.
17. Give the features of a port.
18. Give examples for collaborative groups.
19. Differentiate between configurations and modules in
TinyOS.
20. Mention the classification of code in TinyOS.
72
 PART B
1. Write short notes on (i) Berkley Notes (ii)
Programming Challenges.
2. Explain the various Node level software platforms
available for sensor networks.
3. Explain the various Node level simulators available for
sensor networks.
4. Discuss the concept of centric programming and its
collaborative groups with relevant examples.

73