CHAPTER: 1

INTRODUCTION

1.1 Overview :
Wireless sensor networks consist of individual nodes that are able to interact with the
environment by sensing or controlling physical parameters. These nodes have to
collaborate to fulfill their tasks. The nodes are interlinked together and by using
wireless links each node is able to communicate and collaborate with each other [1].

A Wireless Sensor Network (WSN) is a collection of sensors with limited resources

that collaborate in order to achieve a common goal. Sensor nodes operate in hostile
environments such as battle fields and surveillance zones [1]. Due to their operating
nature, WSNs are often unattended, hence prone to several kinds of novel attacks.
WSNs have attracted a lot of attention recently due to their broad applications in both
military and civilian operations. Many WSNs are deployed in unattended and often
hostile environments such as military and homeland security operations. Therefore,
security mechanisms providing congeniality, authentication, data integrity, and non-
repudiation, among other security objectives, are vital to ensure proper network
operations.
A future WSN is expected to consist of hundreds or even thousands of sensor nodes.
This renders it impractical to monitor and protect each individual node from either
physical or logical attack. It is also unrealistic and uneconomical to enclose each node
in tamper-resistant hardware. Thus, each node represents a potential point of
compromise [1]. Once com-promising certain nodes and acquiring their keying
material adversaries can launch various insider attacks.

1.2 Architecture Of WSN:

For example, they might spoof, alter or replay routing information to interrupt the
network routing [1].
As shown in Figure 1.1, the wireless sensor network and the classical infrastructure
comprises of the standard components like sensor nodes (used as source,
sink/actuators), gateways, Internet, and satellite link, etc.

1|Page
1.2.1 Sensor nodes:
Sensor nodes are the network components that will be sensing and delivering the data.
Depending on the routing algorithms used, sensor nodes will initiate transmission
according to measures and/or a query originated from the Task Manager. According
to the system application requirements, nodes may do some computations [2]. After
computations, it can pass its data to its neighboring nodes or simply pass the data as it
is to the Task Manager. The sensor node can act as a source or sink/actuator in the
sensor field. The definition of a source is to sense and deliver the desired information.
Hence, a source reports the state of the environment. On the other hand, a
sink/actuator is a node that is interested in some information a sensor in the network
might be able to deliver. [2]

Figure 1.1: Illustration of sensor network and backbone infrastructure.

1.2.2 Gateways:
Gateways allow the scientists/system managers to interface Motes to personal
computers (PCs), personal digital assistants (PDAs), Internet and existing networks

2|Page
and protocols. In a nutshell, gateways act as a proxy for the sensor network on the
Internet. According to [2], gateways can be classified as active, passive, and hybrid.
Active gateway allows the sensor nodes to actively send its data to the gateway
server. Passive gateway operates by sending a request to sensor nodes. Hybrid
gateway combines capabilities of the active and passive gateways.

1.2.3 Task Managers:

The Task Manager will connect to the gateways via some media like Internet or
satellite link [3]. Task Managers comprise of data service and client data browsing
and processing. These Task Managers can be visualized as the information retrieval
and processing platform. All information (raw, filtered, processed) data coming from
sensor nodes is stored in the task managers for analysis. Users can use any display
interface (i.e. PDA, computers) to retrieve/analyze these information locally or
remotely.

1.3 System Components and Operations In A Wireless Sensor

Network:

1.3.1 Communication Architecture:

In this section, we will explore the left black box in Figure 1.1, i.e. the sensor field.
The components and operations between sensor nodes within the sensor field would
be explored. We first describe the wireless sensor network architecture and the
communication protocols for the wireless sensor network. This is essential to
understand the hardware and software level power savings strategies. One of the
intension of this report is to provide a survey of the sensor nodes in literature and
recommend the appropriate hardware based on the specific application. We can refer
to [4-7] for more information in the detail composite of the hardware.

1.3.2 Sensor Node:

As mentioned earlier, the sensor field constitutes sensor nodes. Typically, a sensor
node can perform tasks like computation of data, storage of data, communication of

3|Page
data and sensing/actuation of data. A basic sensor node typically comprises of five
main components and they are namely controller, memory, sensors and actuators,
communication device and power supply (see Figure 1.2). A controller is to process
all the relevant data, capable of executing arbitrary code. Memory is used to store
programs and intermediate data. Sensors and actuators are the actual interface to the
physical world. These devices observe or control physical parameters of the
environment. The communication device sends and receives information over a
wireless channel. And finally, the power supply is necessary to provide energy. In
wireless sensor networks, power consumption efficiency is one of the most important
design considerations. Therefore, these intertwined components have to operate and
balance the trade-offs between as small energy consumption as possible and also the
need to fulfill their tasks.

Figure 1.2: Overview of sensor node hardware component.

1.3.2.1 Controller:
Microcontrollers used in several wireless sensor node prototypes are Atmel processor
and Intel Armstrong processors, etc. In this project, we have consolidated a list of
sensor nodes in the literature. It is noted that mica 2 mote and mica Z mote, and mica
2 dot motes are appropriate nodes suitable for large area wetland monitoring
application because of its characteristics. These three motes operation range can out
reached up to 500 feet (152 m), and has the lifetime up to 7 years.

4|Page
1.3.2.2 Communication Device:
Communication device is used to exchange data between individual nodes. The
communication medium between the two nodes is through radio frequencies (wireless
medium). Radio frequency-based communication fits the requirements of most
wireless sensor applications because it provides relatively long range and high data
rates, acceptable error rates at reasonable energy expenditure, and does not require
line of sight between sender and receiver. The 915 MHz and 2.4 GHz industrial,
scientific and medical (ISM) band has been widely suggested for sensor networks [8].
For actual communication, both a transmitter and a receiver are required in a sensor
node. The essential task is to convert a bit stream coming from a microcontroller (or a
sequence of bytes or frames) and convert them to and from radio waves. As half
duplex operation is recommended in wireless sensor network [8], a transceiver is
generally used. In the transceiver, circuitry includes modulation, demodulation,
amplifiers, filters, mixers. The table below summarizes the frequency bands,
modulation and data parameters that could be used in the communication medium.
The transceiver must provide an interface that allows the medium access control
(MAC) layer to initiate frame transmissions and to hand over the packet from the
main memory of the sensor node into the transceiver (or a byte or a bit stream, with
additional processing required on the micro controller). In other direction, incoming
packets must be streamed into buffers accessible by MAC protocol.

1.4 Characteristics of WSN and Its Challenges:

The characteristics of sensor networks and application requirements have a great

impact on the network design objectives in term of network capabilities and network
performance [9].

1.4.1 Characteristics about WSN:

As compared to the traditional wireless communication networks such as mobile ad
hoc network (MANET) and cellular systems, wireless sensor networks have the
following different characteristics and constraints:

5|Page
1.4.1.1 Battery-Powered Sensor Nodes:
Sensor nodes are mostly powered by battery and are deployed in an environment
where it is very difficult to change or recharge the batteries.

1.4.1.2 Self-Configurable:
Sensor nodes are usually randomly deployed and autonomously configure themselves
into a communication network.

1.4.1.3 Unreliable Sensor Nodes:

Because sensor nodes are prone to physical damages or failures due to its deployment
in harsh or hostile environment.

1.4.1.4 Storage Constraints, Severe Energy and Computation:

Sensors nodes are havinghighly limited energy, computation, and storage capabilities.

1.4.1.5 Redundancy in data :

In most sensor network application, sensor nodes are densely deployed in a region of
interest and collaborate to accomplish a common sensing task. So there is some
redundancy present in data.

1.5 CHALLENGES FACED BY WSN:

The nature of large, ad-hoc, wireless sensor networks presents significant challenges
in designing security schemes. A wireless sensor network is a special network which
has many constraint compared to a traditional computer network.

1.5.1 Wireless Medium:

The wireless medium is inherently less secure because its broadcast nature makes
eavesdropping simple. Any transmission can easily be intercepted, altered, or
replayed by an adversary. The wireless medium allows an attacker to easily intercept
valid packets and easily inject malicious ones. Although this problem is not unique to

6|Page
sensor networks, traditional solutions must be adapted to efficiently execute on sensor
networks. [10]

1.5.2 Ad-Hoc Deployment:

The ad-hoc nature of sensor networks means no structure can be statically defined.
The network topology is always subject to changes due to node failure, addition, or
mobility. Nodes may be deployed by airdrop, so nothing is known of the topology
prior to deployment. Since nodes may fail or be replaced the network must support
self-configuration. Security schemes must be able to operate within this dynamic
environment.

1.5.3 Hostile Environment:

The next challenging factor is the hostile environment in which sensor nodes function.
Motes face the possibility of destruction or capture by attackers. Since nodes may be
in a hostile environment, attackers can easily gain physical access to the devices.
Attackers may capture a node, physically disassemble it, and extract from it valuable
information (e.g. cryptographic keys). The highly hostile environment represents a
serious challenge for security researchers.

1.5.4 Resource Scarcity:

The extreme resource limitations of sensor devices pose considerable challenges to
resource-hungry security mechanisms. The hardware constraints necessitate extremely
efficient security algorithms in terms of bandwidth,computational complexity, and
memory. This is no trivialtask. Energy is the most precious resource for
sensornetworks. Communication is especially expensive in terms ofpower. Clearly,
security mechanisms must give special effortto be communication efficient in order to
be energy efficient[11].

1.5.5 Immense Scale:

The proposed scale of sensor networks poses a significantchallenge for security
mechanisms. Simply networking tensto hundreds of thousands of nodes has proven to
be asubstantial task. Providing security over such a network isequally challenging.

7|Page
Security mechanisms must be scalableto very large networks while maintaining high
computationand communication efficiency.

1.5.6 Unreliable Communication:

Certainly, unreliable communication is another threat to sensor security. The security
of the network relies heavily on a defined protocol, which in turn depends on
communication[11].

1.5.6.1 Unreliable Transfer:

Normally the packet-based routing of the sensor network is connectionless and
thus inherently unreliable.

1.5.6.2 Conflicts :
Even if the channel is reliable, the communication may still be unreliable. This
is due to the broadcast nature of the wireless sensor network.

1.5.6.3 Latency:
The multi-hop routing, network congestion and node processing can lead to
greater latency in the network, thus making it difficult to achieve
synchronization among sensor nodes.

1.5.6.4 Unattended Operation:

Depending on the function of the particular sensor network, the sensor nodes may
be left unattended for long periods of time. There are three main cautions to
unattended sensor nodes [11].

Exposure to Physical Attacks:The sensor may be deployed in an environment open

to adversaries, bad weather, and so on. The probability that a sensor suffers a physical
attack in such an environment is therefore much higher than the typical PCs, which is
located in a secure place and mainly faces attacks from a network.

8|Page
Managed Remotely:Remote management of a sensor network makes it virtually
impossible to detect physical tampering and physical maintenance issues.

No Central Management Point: A sensor network should be a distributed network

without a central management point. This will increase the vitality of the sensor
network. However, if designed incorrectly, it will make the network organization
difficult, inefficient, and fragile. Perhaps most importantly, the longer that a sensor is
left unattended the more likely that an adversary has compromised the node.

1.6 Important Attacks In WSN:

Though there are various attacks in Wireless Sensor Networks, but certain active
attacks,that can be detected with our proposed model are as follows:

1.6.1 Attacks on Information in transit:

In a sensor network, sensors monitor the changes of specific parameters or values and
report to the sink according to the requirement. While sending the report, the
information in transit may be altered, spoofed, replayed again or vanished. As
wireless communication is vulnerable to eavesdropping, any attacker can monitor the
traffic flow and get into action to interrupt, intercept, modify or fabricate [12] packets
thus, provide wrong information to the base stations or sinks. As sensor nodes
typically have short range of transmission and scarce resource, an attacker with high
processing power and larger communication range could attack several sensors at the
same time to modify the actual information during transmission.

1.6.1.1 Black hole Attack:

In this attack, a malicious node acts as a blackhole [13] to attract all the traffic in the
sensor network. Especially in a flooding based protocol, the attacker listens to
requests for routes then replies to the target nodes that it contains the high quality or
shortest path to the base station. Once the malicious device has been able to insert
itself between the communicating nodes (for example, sink and sensor node), it is able
to do anything with the packets passing between them.In fact, this attack can affect

9|Page
even the nodes those are considerably far from the base stations. The black hole attack
is one of the simplest routing attacks in WSNs. In a black hole attack, the attacker
swallows (i.e. receives but does not forward) all the messages he receives, just as a
black hole absorbing everything passing by. By refusing to forward any message he
receives, the attacker will affect all the traffic flowing through it. Hence, the
throughput of a subset of nodes, especially the neighboring nodes around the attacker
and with traffic through it, is dramatically decreased. Different locations of the
attacker induce different influences on the network. If the attacker is located close to
the base station, all the traffic going to the base station might need to go through the
attacker. Obviously, black hole attacks in this case can break the communication
between the base station and the rest of the WSN, and effectively prevent the WSN
from serving its purposes. In contrast, if a black hole attacking node is at the edge of
the WSN, probably very few sensors need it to communicate with others. Therefore,
the harm can be very limited.

1.6.1.2 Sinkhole Attack:

Sinkhole is a more complex attack [14] compared with black hole attack. Given
certain knowledge of the routing protocol in use, the attacker tries to attract the traffic
from a particular region through it. For example, the attacker can announce a false
optimal path by advertising attractive power, bandwidth, or high quality routes to a
particular region. Other nodes will then consider the path through this attacker node
better than the currently used one, and move their traffic onto it. Since affected nodes
depend on the attacker for their communication, the sinkhole attack can make other
attacks efficient by positioning the attacker in busy information traffic. Many other
attacks, such as eavesdropping, selective forwarding and black holes, etc., can be
empowered by sinkhole attacks.

1.6.1.3 Hello Flood Attack

Hello Flood Attack is introduced in [15]. This attack uses HELLO packets as a
weapon to convince the sensors in WSN. In this sort of attack an attacker with a high
radio transmission (termed as a laptop-class attacker in) range and processing power
sends HELLO packets to a number of sensor nodes which are dispersed in a large area

10 | P a g e
within a WSN. The sensors are thus persuaded that the adversary is their neighbor.
As a consequence, while sending the information to the base station, the victim nodes
try to go through the attacker as they know that it is their neighbor and are ultimately
spoofed by the attacker.

1.6.1.4 Wormhole Attack

A wormhole attack [16] requires two or more adversaries. These adversaries

havebetter communication resources (e.g. power, bandwidth) than normal nodes,
andcan establish better communication channels (called “tunnels”) between them.
Unlikemany other attacks in the network layer, the channels are real. Other
sensorsprobably end up adopting the tunnels into their communication paths,
renderingtheir output under the scrutiny of the adversaries.

1.6.1.5 Clone Attack

Clone attack also known as node replication attack, is a severe attack in WSNs. In this
attack, an adversary (WSN Adversarycan be person or another entity that only
monitors the communicationchannels which threatens the confidentiality of data)
captures a few of nodes, replicates them and then deploys arbitrary number of replicas
throughout the network. In clone attack, an adversary may capture a sensor node and
copy the cryptographic information to another node known as cloned node. Then this
cloned sensor node can be installed to capture the information of the network. The
adversary can also inject false information, or manipulate the information passing
through cloned nodes.

Mauro Conti et.al in [17] characterized the clone attack:

1. A clone is considered totally honest by its neighbour. In fact, without global
countermeasures, honest nodes cannot be aware of the fact that they have a
clone among their neighbour.

2. To have a large amount of compromised nodes, the adversary does not need
to compromise a high number of nodes. Indeed, once a single node has been

11 | P a g e
 captured and compromised, the main cost of the attack has been sustained.
Making further clones of the same node can be considered cheap.

1.6.1.6 Denial of Service :

Denial of Service (DOS) [18-19] is produced by the unintentional failure of nodes or
malicious action. The simplest DOS attack tries to exhaust the resources available to
the victim node, by sending extra unnecessary packets and thus prevents legitimate
network users from accessing services or resources to which they are entitled. DOS
attack is meant not only for the adversary’s attempt to subvert, disrupt, or destroy a
network, but also for any event that diminishes a network’s capability to provide a
service.

1.7 TYPES OF DOS ATTACK:

DoS attacks are classified into three groups: collision, unfairness and exhaustion
attacks.
1.7.1 In a collision attack,
the attacker transmits data packets regardless of the status of the broadcast medium.
Such packets collide with the data or control packets from the legitimate sensor nodes.
1.7.2 In an unfairness attack,
the adversary transmits an unusually large number of packets when the medium is
free. This prevents the legitimate sensors from transmitting their packets.
1.7.3 In an exhaustion attack
the adversary transmits an abnormally large number of RTS packets to the normal
sensor nodes, which exhausts them prematurely.
DoS attacks can be detected if abnormal variations occur in sensitive parameters such
as collision rate R; (number of collisions observed by a node per second), average
waiting time Tw (waiting time of a packet in MAC buffer before transmission), and
RTS arrival rate (RRTS) (number of RTS packets received successfully by a node
every second)[20,21].

1.8 SECURITY:
Security is an important aspect of a WSN operation. Many types of attacks on WSNs

12 | P a g e
have been devised and many countermeasures are proposed.
1.8.1 Security Requirements:
The goal of security services in WSNs is to protect the information and resources
from attacks and mis-behaviour. The security requirements in WSNs include:

Availability
which ensures that the desired network services are available even in the presence of
denial-ofservice attacks.

Authorization
which ensures that only authorized sensors can be involved in providing information
to network services.

Authentication,which ensures that the communication from one node to another node
is genuine, that is, a malicious node cannot masquerade as a trusted network node.

Confidentiality, which ensures that a given message cannot be understood by anyone

other than the desired recipients.

Integrity,which ensures that a message sent from one node to another is not modified
by malicious intermediate nodes.

1.8.2 Security Against Dos Attacks:

DoS attacks are classified into three groups: collision, unfairnessand exhaustion
attacks. In a collision attack, the attacker transmits data packets regardless of the
status of the broadcast medium. Such packets collide with the data or control packets
from the legitimate sensor nodes. In an unfairness attack, the adversary transmits an
unusually large number of packets when the medium is free. This prevents the
legitimate sensors from transmitting their packets. In an exhaustion attack, the
adversary transmits an abnormally large number of RTS packets to the normal sensor
nodes, which exhausts them prematurely.
DoS attacks can be detected if abnormal variations occur in sensitive parameters such

13 | P a g e
as collision rate R; (number of collisions observed by a node per second), average
waiting time Tw (waiting time of a packet in MAC buffer before transmission), and
RTS arrival rate (RRTS) (number of RTS packets received successfully by a node
every second)[22-23].

1.9 SECURITY GOALS FOR SENSOR NETWORKS:

As the sensor networks can also operate in an Adhoc manner the security goals cover
both those of thetraditional networks and goals suited to the unique constraints of
Adhoc sensor networks. The security goalsare classified as primary and secondary
[56]. The primarygoals are known as standard security goals such asConfidentiality,
Integrity, Authentication and Availability (CIAA). The secondary goals are Data
Freshness, Self- Organization, Time Synchronization and SecureLocalization.

1.9.1 The primary goals are:

Data Confidentiality:
Confidentiality is the ability to conceal messages froma passive attacker so that any
message communicated viathe sensor network remains confidential. This is the most
important issue in network security. A sensor node shouldnot reveal its data to the
neighbors.

Data Authentication:
Authentication ensures the reliability of the message by identifying its origin. Attacks
in sensor networks do notjust involve the alteration of packets; adversaries can also
inject additional false packets [57]. Data authenticationverifies the identity of the
senders and receivers. Data authentication is achieved through symmetric or
asymmetric mechanisms where sending and receiving nodes share secret keys. Due to
the wireless nature of themedia and the unattended nature of sensor networks, it
isextremely challenging to ensure authentication.

Data Integrity:
Data integrity in sensor networks is needed to ensurethe reliability of the data and

14 | P a g e
refers to the ability toconfirm that a message has not been tampered with, altered or
changed. Even if the network has confidentialitymeasures, there is still a possibility
that the data integrityhas been compromised by alterations.
The integrity of thenetwork will be in trouble when:
 A malicious node present in the network injectsfalse data.
 Unstable conditions due to wireless channel causedamage or loss of data.[55]

Data Availability:
Availability determines whether a node has the ability to use the resources and
whether the network is available forthe messages to communicate. However, failure
of the base station or cluster leader’s availability will eventuallythreaten the entire
sensor network. Thus availability is ofprimary importance for maintaining an
operational network.

1.9.2 The Secondary goals are:

Data Freshness:
Even if confidentiality and data integrity are assured, there is a need to ensure the
freshness of each message.Informally, data freshness [55] suggests that the data is
recent, and it ensures that no old messages have beenreplayed. To solve this problem
a nonce, or another time related counter, can be added into the packet to ensure
datafreshness.

Self-Organization:
A wireless sensor network is a typically an ad hoc network, which requires every
sensor node be independentand flexible enough to be self-organizing and self-healing
according to different situations. There is no fixedinfrastructure available for the
purpose of network management in a sensor network. This inherent featurebrings a
great challenge to wireless sensor network security. If self-organization is lacking in a
sensor network, the damage resulting from an attack or even the riskyenvironment
may be devastating.

15 | P a g e
Time Synchronization:
Most sensor network applications rely on some form oftime synchronization.
Furthermore, sensors may wish tocompute the end-to-end delay of a packet as it
travelsbetween two pairwise sensors. A more collaborativesensor network may
require group synchronization [55] for tracking applications.

Secure Localization:
Often, the utility of a sensor network will rely on its ability to accurately and
automatically locate each sensorin the network. A sensor network designed to locate
faults will need accurate location information in order topinpoint the location of a
fault. Unfortunately, an attacker can easily manipulate non secured location
information by reporting false signal strengths, replaying signals.

1.10 APPLICATIONS OF WSN:

Wireless Sensor Networks (WSN) has off late, found applications in wide-ranging
areas. In this section we list some of the prominent areas of applications of WSN. The
list would be very lengthy if we exhaust all the areas of WSN applications.
Therefore, in this paper only handful applications are provided.
The military applications of sensor nodes include battlefield surveillance and
monitoring, guiding systems ofintelligent missiles and detection of attack by
weapons of massdestruction.
The Medical Application: Sensors can be extremely useful in patient
diagnosis and monitoring [58]. Patientscan wear small sensor devices that
monitor their physiologicaldata such as heart rate or blood pressure.
Environmental monitoring: It includes traffic,habitat, Wild fire etc.

Industrial Applications: It includes industrial sensing and diagnostics. For

example appliances, factory, supplychains etc.

CHAPTER: 2
LITERATURE SURVEY

16 | P a g e
RishavDubey, Vikram Jain, Rohit Singh Thakur, SiddharthDuttChoubey in
(2012) proposed “Attacks in Wireless Sensor Networks”.[24]

The authors proposed that Wireless Sensor Networks is an emerging technology.

WSN has limitations of system resources like battery power, communication range
and processing capability. WSNs are used in many applications in military,
ecological, and health-related areas. These applications often include the monitoring
of sensitive information such as enemy movement on the battlefield or the location of
personnel in a building. One of the major challenges wireless sensor networks face
today is security, so there is the need for effective security mechanism .In this
research they investigate how wireless sensor networks can be attacked in practice.

Rajkumar, Sunitha K.R and Dr. H.G Chandrakanth (2012) surveyed on“A
Survey on Security Attacks in Wireless Sensor Network”.[25]
A wireless sensor network (WSN) has important applications such as remote
environmental monitoring and target tracking. This has been enabled by the
availability, particularly inrecent years, of sensors that are smaller, cheaper, and
intelligent. These sensors are equipped with wireless interfaces with which they can
communicate with one another to form a network. In this paper we deal with the
security of the wireless sensor networks. Staring with a brief overview of the sensor
networks, and discusses the current state of the security attacks in WSNs. Various
types of attacks are discussed and their countermeasures presented. A brief discussion
on the future direction of research in WSN security is also included.

WazirZada Khan Yang Xiang Mohammed Y Aalsalem, in (2011) proposed

“Comprehensive Study of Selective Forwarding Attack in Wireless Sensor
Networks”.[26]
Sensor networks are becoming closer towards wide-spread deployment so security
issues become a vital concern. Selective forwarding attack is one of the harmful
attacks against sensor networks and can affect the whole sensor network

17 | P a g e
communication. The variety of defense approaches against selective forwarding attack
is overwhelming. In this research they had described all the existing defensive
schemes according to our best knowledge against this attack along with their
drawbacks, thus providing researchers a better understanding of the attack and current
solution space. Also classifies proposed schemes according to their nature and
defense. Nature of scheme classifies into Distributed and Centralized. Defense of
scheme classifies into detection and prevention.

Chaudhari H.C. and Kadam L.U. (2011) research on “Wireless Sensor

Networks: Security, Attacks and Challenges”.[27]
The significant advances of hardware manufacturing technology and the development
of efficient software algorithms make technically and economically feasible a network
composed of numerous, small, low-cost sensors using wireless communications, that
is, a wireless sensor network. Sensor networks have great potential to be employed in
mission critical situations like battlefields but also in more everyday security and
commercial applications such as building and traffic surveillance, habitat monitoring
and smart homes etc. However, wireless sensor networks pose unique security
challenges. Security is becoming a major concern for WSN protocol designers
because of the wide security-critical applications of WSNs .In this paper we have
made an effort to document all the known security issues in wireless sensor networks
and discusses a wide variety of attacks in WSN and their classification mechanisms
and different securities available to handle them including the challenges faced. In
this research they took up the challenge and have proposed an integrated
comprehensive security that will provide security services for all services of sensor
network. The sensing technology combined with processing power and wireless
communication makes it profitable for being exploited in great quantity in future. The
wireless communication technology also acquires various types of security threats.

Jiyong Son , Seoul and Hwan-JooKwak, 2011 introduced “Back propagation

neural network based real-time self-collision detection method”. [28]
This research proposed a back propagation neural network based real-time humanoid
self-collision detection method which eliminates the repetition of detection
computation for same and similar motion sets. The proposed system is able to reduce

18 | P a g e
self-collision detection computation time significantly, because of the pattern
recognition capability of the neural network. However, the accuracy of back
propagation neural network based self-collision detection cannot be guaranteed 100%.
For this reason, the system is also designed to detect potential miss detected motion
sets though the module based self-collision detection method, which eliminates
unnecessary motion pairs by focusing on certain modules with higher collision
probability. This module based self-collision detection method is a failsafe method.
Kalpana Sharma, M K Ghose (2010) worked on “Wireless Sensor Networks: An
Overview on its Security Threats”. [29]Wireless sensor networks have become a
growing area of research and development due to the tremendous number of
applications that can greatly benefit from such systems and has lead to the
development of tiny, cheap, disposable and self-contained battery powered computers,
known as sensor nodes or “motes”, which can accept input from an attached sensor,
process this input data and transmit the results wirelessly to the transit network.
Despite making such sensor networks possible, the very wireless nature of the sensors
presents a number of security threats when deployed for certain applications like
military ,surveillances etc. The problem of security is due to the wireless nature of the
sensor networks and constrained nature of resources on the wireless sensor nodes,
which means that security architectures used for traditional wireless networks are not
viable. Furthermore, wireless sensor networks have an additional vulnerability
because nodes are often placed in a hostile or dangerous environment where they are
not physically protected. In this research the authors discussed some security threats
and challenges faced by WSNs.

Dr. G. Padmavathi and Mrs. D. Shanmugapriya, in (2009) reviewed on “A

Survey of Attacks, Security Mechanisms and Challenges in Wireless Sensor
Networks”. [30]

Wireless Sensor networks (WSN) is an emerging technology and have great potential
to be employed in critical situations like battlefields and commercial applications such
as building, traffic surveillance, habitat monitoring and smart homes and many more
scenarios. One of the major challenges wireless sensor networks face today is

19 | P a g e
security. While the deployment of sensor nodes in an unattended environment makes
the networks vulnerable to a variety of potential attacks, the inherent power and
memory limitations of sensor nodes makes conventional security solutions unfeasible.
The sensing technology combined with processing power and wireless
communication makes it profitable for being exploited in great quantity in future. The
wireless communication technology also acquires various types of security threats.
This research discussed a wide variety of attacks in WSN and their classification
mechanisms and different securities available to handle them including the challenges
faced.

Ahmad Hosseingholizadeh, Dr. AbdolrezaAbhari,(2009) proposed “A neural

network approach for Wireless sensor network power management”. [31]

In this research they introduced a new approach for wireless sensor network power
management which is based on neural networks. In this new approach an intelligent
analysis is used to process the structure of a wireless sensor network (WSN) and
produce some information which can be used to improve the performance of WSNs’
management application. In this paper we introduce our neural network based
approach which results in a more efficient way.

Hong-Ning Dai in 2009 proposed “Throughput and Delay in Wireless Sensor

Networks”. [32]

Author proposed that wireless sensor networks are equipped with only omni-
directional antennas, which can cause high collisions. It is shown that the per node
throughput in such networks is decreased with the increased number of nodes. Thus,
the transmission with multiple short range hops is preferred to reduce the interference.
However, other studies show that the transmission delay increases with the increased
number of hops.

Raghavendra V. Kulkarni, 2009 worked on “Neural Network Based Secure

Media Access Control Protocol for Wireless Sensor Networks”. [33]

20 | P a g e
This research discussed an application of a neural network in wireless sensor network
security. It presents a neural network approach against the denial-of-service attacks
launched by adversaries. The MLP enhances the security of a WSN by constantly
monitoring the parameters that exhibit unusual variations in case of an
attack/collision. The MLP shuts down the MAC layer and the physical layer of the
sensor node when the suspicion factor, the output of the MLP, exceeds a preset
threshold level. Backpropagation and particle swarm optimization algorithms are used
to avoid the attack.

David Boyle, 2008 introduced “Securing Wireless Sensor Networks: Security

Architectures”. [34]

Wireless sensor networking remains one of the most exciting and challenging
research domains of our time. As technology progresses, so do the capabilities of
sensor networks. Limited only by what can be technologically sensed, it is envisaged
that wireless sensor networks will play an important part in our daily lives in the
foreseeable future. Privy to many types of sensitive information, both sensed and
disseminated, there is a critical need for security in a number of applications related to
this technology. Resulting from the continuous debate over the most effective means
of securing wireless sensor networks, this research considered a number of the
security architectures employed, and proposed, to date, with this goal in sight. They
are presented such that the various characteristics of each protocol are easily
identifiable to potential network designers, allowing a more informed decision to be
made when implementing a security protocol for their intended application.
Authentication is the primary focus, as the most malicious attacks on a network are
the work of imposters, such as DOS attacks, packet insertion etc. Authentication can
be defined as a security mechanism, whereby, the identity of a node in the network
can be identified as a valid node of the network. Subsequently, data authenticity can
be achieved; once the integrity of the message sender/receiver has been established.

Phillip Reindl and Kendall Nygard, 2007 proposed “Defending Malicious

Collision Attacks in Wireless Sensor Networks”. [35]

21 | P a g e
Authors discussed that Security is an important issue for sensor networks deployed in
hostile environments, such as military battlefields. The low cost requirement
precludes the use of tamper resistant hardware on tiny sensor nodes. Hence, sensor
nodes deployed in open areas can be compromised and used to carry out various
attacks on the network. In this paper, we consider the collision attack that can be
easily launched by a compromised (or hostile) node: a compromised node does not
follow the medium access control protocol and cause collisions with neighbor
transmissions by sending a short noise packet. This attack does not consume much
energy of the attacker but can cause a lot of disruptions to the network operation. Due
to the wireless broadcast nature, it is not trivial to identify the attacker. In this paper,
we propose a distributed scheme that is based on low-cost hardware and can
effectively identify the source of a collision attack.

Frank Oldewurtel in 2005 proposed “Neural Wireless Sensor Networks”. [36]

They presented an overview of embedded network applications and discuss

requirements arising from this analysis. Furthermore, they discussed selected in-
network processing techniques and point out the analogy between neural and sensor
networks. In the following neural networks are introduced in the sensor network
context. They described the motivation and the practical case for neural networks in
the sensor networks context, and evaluate early results achieved with our test
implementation. They argued that there is a high potential with these paradigms
which promise a strong impact on the future research, especially if applied as a hybrid
technology using Hopfield neural network.

Kai Xing, ShyaamSundhar and RajamadamSrinivasan, 2005 introduced

“Attacks and Countermeasures in Sensor Networks: A Survey”. [37]
Author proposed that a wireless sensor network (WSN) is comprised of a large
number of sensors that collaboratively monitor various environments. The sensors all
together provide views of the environments that offer more information than those
local views provided by independently operating sensors. There are numerous
potential applications of WSNs in various areas such as residence, industry, military

22 | P a g e
and many others. For instance, people can use WSNs to build intelligent house, to
gather machine information for real-time control in factories, and to track enemy
movements in battle fields. To collect data from WSNs, base stations and aggregation
points are commonly used. They usually have more resources (e.g. computation
power and energy) than normal sensor nodes which have more or less such
constraints. Aggregation points gather data from nearby sensors, integrate the data
and forward them to base stations, where the data are further processed or forwarded
to a processing center. In this way, energy can be conserved in WSNs and network
life time is thus prolonged.

K. Jamieson, H. Balakrishnan, (2003) proposed “A mac protocol for event-

driven wireless sensor networks”. [38]
This research demonstrates that CSMA based protocols cannot avoid such collisions
when the number of concurrent transmissions grows. Such collisions become severe
in dense WSNs for two reasons. First, many dense sensor networks are event-driven
and generate bursty spatially-correlated traffic, where multiple sensors in the same
neighborhood all have messages to send simultaneously in response to the same
event. Second, WSNs are typically equipped with few sinks to which packets from
sensors converge.

Simon X. Yang, 2001 “Neural Network Approaches to Dynamic Collision-Free

Trajectory Generation”. [39]
In this research, dynamic collision-free trajectory generation in a non-stationary
environment is studied using biologically inspired neural network approaches. The
proposed neural network is topologically organized, where the dynamics of each
neuron is characterized by a neural network. The effectiveness and efficiency of the
proposed approaches are demonstrated through simulation and comparison studies.

CHAPTER: 3

23 | P a g e
 PROBLEM FORMULATION AND METHODOLOGY

3.1 How Collision Occur in WSN:

Collision occurs when two or more nodes attempt to transmit a packet across the
network at the same time. The transmitted packets must be discarded and then
retransmitted, thus the retransmission of those packets increases the energy
consumption and the latency. Collision attack is a type of DOS attack which occurs
on Data Link Layer. Packet Collision occurs when two or more close stations attempt
to transmit a packet at the same time. This can result in packet loss and impede
network performance. Many CSMA based MAC protocols are proposed in Wireless
Sensor Network (WSNs) to avoid collisions, such as B-MAC [40]. These protocols
can efficiently reduce collisions, but intrinsically cannot eliminate all collisions,
because of hidden terminal problems, as well as collisions when multiple nodes sense
the medium free at the same time. Furthermore, the consequences of packet collisions
are serious to WSNs. Collisions can cause the loss of critical control information from
base stations, and applications may fail.

3.2 Role of Neural Network in WSN:

Although neural network and sensor network are normally viewed as two radically
different subjects, they do share one thing in common. The most fundamental way of
exchanging information in both kinds of networks is one-to-many communication,
i.e., the broadcast. In a biological neural network, a firing neuron sends an action
potential to all neurons that are connected to it by synapses, each of which may
impose different delay and amplification to the transmitted signal. Similarly, a
communication node in a sensor network broadcasts its signal to all nodes within its
transmission range. The proposed computing with time paradigm applies to networks
in which a broadcast is a Communication primitive, such as neural networks in
biology or wireless networks in telecommunication. Another example of such a
paradigm is computing with action potentials proposed by Hopfield et al. [41], who
observed that analog information can be encoded into firing times of action potentials
and that the timings of these action potentials can be used to carry out a vector
matching algorithm. The ability to perform broadcast-based communication was not

24 | P a g e
explicitly mentioned as a requirement. There must be certain moments at which
distributed neurons observe the same events, as if each of them would own a local
clock and these clocks were synchronized from time to time by such events.
Broadcast naturally provides plenty of such synchronization points. The purpose is to
look for an optimum value by associating the firing times with a certain variable in
such a way that the smaller this variable is the more desirable the property of the
corresponding neuron is. Hence, the neuron firing earliest will naturally be the one
whose property variable has the minimum value among the neurons being compared.
Hence, the essence here is to introduce competition, instead of superposition in
Hopfield’s approach, to select a winner that possesses the desired optimality.

3.3 Feed Forward Back Propagation:

ANN’s are biologically inspired computer programs to simulate the way in which the
human brain process information. It is a very powerful approach for building complex
and nonlinear relationship between a set of input and output data. The power of
computation comes from connection in a network. Each neuron has weighted inputs,
simulation function, transfer function and output. The weighted sum of inputs
constitutes the activation function of the neurons. The activation signal is passed
through a transfer function which introduces non-linearity and produces the output.
During training process, the inter-unit connections are optimized. Once the network is
trained, new unseen input information is entered to the network to calculate the test
output. There are many back propagation algorithm are used in the neural network but
mostly used feed forward back propagation neural network (FBNN).

Figure 3.1: Simple two layer feed forward back propagation neural network

A two layer simple Feed-forward back-propagation neural networks consist of

25 | P a g e
basically three layers: an input layer, hidden layer and output layer.

3.4 Hopfield Neural Network:

The Hopfield neural network is a simple artificial network which is able to store
certain memories or patterns. Hopfieldneural network model is a fully interconnected
network of binary units with symmetric connection weights between the units.The
nodes in the network are vast simplifications of real neurons - they can only exist in
one of two possible states - firing or not firing. At any instant of time a node will
change its state depending on the inputs it receives from itself and the other nodes.
The dynamics of the Hopfield network can be described formally in mathematical
terms. The activation levels of binary units are set to zero and one for "off" and "on,"
respectively. Starting from some initial configuration (V0,V1,V2…Vi) where is number
of units and is the activation level of unit. The behavior of network is determined by
an appropriate energy function. This function is based on neuron states, weights and
bias value derived from problem data. Update rule of neurons is defined based on
energy function [42]

3.4.1 Hopfield network applied to the single sensor node:

Wireless communication often suffers from bad channel conditions. One has to deal
with erroneous or even lost data packets by signal processing algorithms or other
techniquessuch as retransmission. In this context the HN shows promising
characteristics such as associative memory, robustness and error correction capability
to overcome this drawback [43-45]. Associative memory means that a pattern is not
stored on any individual neuron but is a property of the whole network. Thus, the
weights within the HN store the average correlations between all pattern components
across all patterns. The network presented with a partial or corrupted pattern can then
use the correlations to recreate the entire pattern. The HN itself is robust since it
performs pattern completion in case of missing data and pattern correction in case of
corrupted datadue to the association ability. The HN is a single layer fully connected
feedback network with no direct feedback connections, i.e. each single neuron is not
directly connected to itself. Furthermore, it shows symmetric (bidirectional) weights,
i.e. the weights between all single neurons are equal in either direction. Figure 3.2

26 | P a g e
depicts the HN, which is presented with the sensor input pattern containing readings
of three sensors. After iterative processing the optimized, i.e. completed or corrected
pattern can be used to build a data packet, which is represented by the dashed box.

Figure3.2: The Hopfield network applied to the single sensor node.

3.5 Collision Avoidance Using HNN:

Wireless communication often suffers from bad channel conditions. One has to deal
with erroneous or even lost data packets by signal processing algorithms or other
techniques such as retransmission. In this context the HNN shows promising
characteristics such as associative memory, robustness and error correction capability
to overcome this drawback. HNN uses the concepts of associative memory, pattern
completion and error correction. Here Associative memory means that a pattern is not
stored on any individual neuron but is a property of the whole network. By using
partial or corrupted pattern, it can then use the correlations to recreate the whole
pattern/network then it performs pattern /network completion for whole network.
The HN itself is robust in case of missing data, and pattern correction in case of
corrupted data to remove the collided packet from the network due to the association
ability. The HN is a single layer fully connected feedback network with no direct
feedback connections, i.e. each single neuron is not directly connected to itself[46-
48].

27 | P a g e
3.6 Proposed Algorithm:
1. Create a network having 18 node arranged in circular fashion.
2. Select source and destination and the sensor node from the nodes.
3. While(data is not received by destination)
4. repeat
a. If(sensor node detect collision)
b. Then
i. Apply pattern recognition neural network to change the
position of the node at which collision is detected.
ii. And start transmission from source node again.
c. Else
i. Transmit the data from one node to another.
5. End if
6. End while
7. Exit

3.7 TOOL DISCRIPTION:

3.7.1 INTRODUCTION:
The name MATLAB stands for MATrixLABoratory. MATLAB was written
originally to provide easy access to matrix software developed by the LINPACK
(linear system package) and EISPACK (Eigen system package) projects. MATLAB
[49] is a high-performance language for technical computing. It integrates
computation, visualization, and programming environment. Furthermore, MATLAB
is a modern programming language environment: it has sophisticated data structures,
contains built-in editing and debugging tools, and supports object-oriented
programming. These factors make MATLAB an excellent tool for teaching and
research.
MATLAB has many advantages compared to conventional computer languages (e.g.,
C, FORTRAN) for solving technical problems. MATLAB is an interactive system
whose basic data element is an array that does not require dimensioning. The software
package has been commercially available since 1984 and is now considered as a
standard tool at most universities and industries worldwide [50].

28 | P a g e
3.7.2 HISTORY OF MATLAB:
Cleve Moler, the chairman of the computer science department at the University of
New Mexico, started developing MATLAB in the late 1970s. He designed it to give
his student’s access to LINPACK and EISPACK without them having to
learn FORTRAN. It soon spread to other universities and found a strong audience
within the applied mathematics community. Jack Little, an engineer, was exposed to it
during a visit Moler made to Stanford University in 1983. Recognizing its commercial
potential, he joined with Moler and Steve Bangert. They rewrote MATLAB in C and
founded  MathWorks in 1984 to continue its development. These rewritten libraries
were known as JACKPAC. In 2000, MATLAB was rewritten to use a newer set of
libraries for matrix manipulation, LAPACK.

MATLAB was first adopted by researchers and practitioners in control engineering,

Little's specialty, but quickly spread to many other domains. It is now also used in
education, in particular the teaching of linear algebra and numerical analysis, and is
popular amongst scientists involved in image processing. [51]

3.7.3 ADVANTAGES OF MATLAB:

 MATLAB may behave as a calculator or as a programming language
 It combines nicely calculation and graphic plotting.
 It is relatively easy to learn.
 It is interpreted (not compiled), errors are easy to fix.
 It is optimized to be relatively fast when performing matrix operations
MATLAB does have some object-oriented element
 It is an interactive system whose basic data element is an array that does not
require dimensioning. The software package has been commercially available
since 1984 and is now considered as a standard tool at most universities and
industries worldwide.
 It has powerful built-in routines that enable a very wide variety of
computations.
 It also has easy to use graphics commands that make the visualization of
results immediately available. Specific applications are collected in packages
referred to as toolbox. There are toolboxes for signal processing, symbolic

29 | P a g e
 computation, control theory, simulation, optimization, and several other fields
of applied science and engineering.

3.7.4 DISADVANTAGES OF MATLAB:

 MATLAB is not a general purpose programming language such as C, C++, or
FORTRAN.
 It is designed for scientific computing, and is not well suitable for other
applications.
 It is an interpreted language, slower than a compiled language such as C++.
 MATLAB commands are specific for MATLAB usage. Most of them do not
have a direct equivalent with other programming language commands.

3.7.5 APPLICATIONS OF MATLAB:

1) Data Exploration, Acquisition, Analyzing & Visualization.

2) Engineering drawing and Scientific graphics.
3) Analyzing of algorithmic designing and development.
4) Mathematical functions and Computational functions.
5) Simulating problems prototyping and modeling.
6) Application development programming using GUI building
environment. [52]

3.7.6 INSTALLATION STEPS OF MATLAB:

Step 1: Start the Installer
The method you use to start the installer depends on your platform.

• Windows — Insert the DVD into the DVD drive connected to your system or
double-click the installer file you downloaded from the Math Works Website [49].
The installer should start automatically.

30 | P a g e
Step 2: Choose to Install Without Using the Internet
If you do not have an Internet connection, select the Install without using the
Internet option and click Next.

Step 3: Review the License Agreement

Review the software license agreement and, if you agree with the terms, select
Yesand click Next.
After the installation is complete, you can view or print the license agreement using
the file license.txt located in the top-level installation folder.

Step 4: Specify the File Installation Key

If you do not have an Internet connection, and choose to install manually, the installer
displays the File Installation Key dialog box. A File Installation Key identifies the
products you can install.

If you have the key, select the I have the File Installation Key for my license
option, enter the File Installation Key, and click Next. Theadministrator contact on a
license can retrieve the File Installation Key fromthe License Center at the Math
Works Web site.

If you do not have the key, select the I do not have the File Installation Key option
and click Next. The installer will provide you with the informationyou need to get a
key.

If You Do Not Have the File Installation Key

The Installation and Activation Next Steps dialog box contains the information you
need to retrieve your File Installation Key from the License Center at the MathWorks
Web site. This information includes:
• Host ID
• Release number (for example, R2012b)
• Operating system user name (Note that user names are case-sensitive in activation.)
To get you’re File Installation Key:

31 | P a g e
 1. Remember the information displayed in this dialog box and click Finish to
exit the installer. On Windows and Linux systems, you can click Print to print
out the information.
2. Go to a computer with an Internet connection and log in to your account at the
Math Works Web site.

3. Visit the License Center and enter the information from this dialog box. Math
Works use information to generate a File Installation Key and License File
for your license.
4. Return to your computer and re-run the installer. With the File Installation
Key and a License File, you can install and activate the software without an
Internet connection.

Step 5: Choose the Installation Type

In the Installation Type dialog box, specify whether you want to perform a Typical or
Custom installation and click Next.

• Choose Typical if you have an Individual or Group license and do not need to
specify which products you want to install and do not need to access any installation
options.

• Choose Custom if you need to specify which products to install, need access to
installation options, or need to install the license manager (network license options
only).

If you choose a Typical installation, the installer skips the product selection and
installation options steps.

Step 6: Specify the Installation Folder [50]

Specify the name of the folder where you want to install Math Works products.
Accept the default installation folder or click Browse to select a different one. If the
folder doesn’t exist, the installer creates it.

32 | P a g e
On Macintosh systems, the installer puts the MATLAB application package,
MATLAB_R2012b.app, in the Applications folder, by default. When specifying a
folder name, you can use any alphanumeric character and some special characters,
such as underscores. The installer tells you if the name you specified includes any
characters that are not permitted in file names. If you make a mistake while entering a
folder name and want to start over, click Restore Default Folder. After making your
selection, click Next

Step 7: Specify Products to Install (Custom Only) [53]

If you are performing a custom installation, you can specify which products you want
to install in the Product Selection dialog box. This dialog box lists all the products
associated with the license you selected or with the Activation Key you specified. In
the dialog box, all the products are preselected for installation. If you do not want to
install a particular product, clear the check box next to its name.

After selecting the products you want to install, click Next to continue with the
installation.

Step 8: Specify Installation Options (Custom Only)

For Custom installations, you can specify several installation options,depending on
your platform. Windows Systems On Windows, the Installation Options dialog box
lets you choose whether to put shortcuts for starting MATLAB software in the Start
menu and on the desktop. After selecting installation options, click Nextto proceed
with the installation.

Step 9: Complete the Installation [54]

When the installation successfully completes, the installer displays the Installation
Complete dialog box. In this dialog box, you can choose to activate the software you
just installed. You cannot use the software you installed until you activate it.
MathWorks recommends activating immediately after installation. Click Next to
proceed with activation.

If you choose to exit the installer without performing activation, clear the Activate

33 | P a g e
MATLAB option and click Finish(the button label changes). You can activate later
using the activation application.

Step 10: Activate Your Installation

Because you were not logged in to your MathWorks Account during installation, or
you started the activation application independently, you must choose whether to
activate automatically or manually. Select the Activate manually without the
Internet option and click Next.

Step 11: Specify the Path to the License File

To activate without an Internet connection, you must have a License File. The License
File identifies which products you can run. The administrator contact on the license
can retrieve the License File from the License Center at the MathWorks Web site.
Select the Enter the path to the License File option and enter the full path of your
License File in the text box (or drag and drop the file) and click Next.

If you do not have your License File, select the I do not have a license file option and
click Next to get information about how to retrieve a License File.
3.7.7 STARTING WITH MATLAB:
After logging into your account, you can enter MATLAB by double-clicking on the
MATLAB shortcut icon (MATLAB 7.0.4) on your Windows desktop. When you start
MATLAB, a special window called the MATLAB desktop appears. The desktop is a
window that contains other windows. The major tools within or accessible from the
desktop are:
 The Command Window
 The Command History
 The Workspace
 The Current Directory
 The Help Browser
 The Start button

When MATLAB is started for the first time, the screen looks like the one that shown

34 | P a g e
in the Figure 3.3.
This illustration also shows the default configuration of the MATLAB desktop. You
can customize the arrangement of tools and documents to suit your needs. Now, we
are interested in doing some simple calculations. We will assume that you have
sufficient understanding of your computer under which MATLAB is being run.
You are now faced with the MATLAB desktop on your computer, which contains the
prompt (>>) in the Command Window. Usually, there are 2 types of prompt:
>> for full version
EDU> for educational version

Fig. 3.3 The Graphical Interface to the MATLAB work

3.7.8 MATHEMATICAL FUNCTIONS:
MATLAB offers many predefined mathematical functions for technical computing
which contains a large set of mathematical functions.
Table 4.1 lists some commonly used functions, where variables x and y can be
numbers, vectors, or matrices.

35 | P a g e
Table:3.1 Elementary Functions used in MATLAB

36 | P a g e
 CHAPTER: 4
RESULT AND DISCUSSION

In this section, the performance of each classifier in terms of packet delivery ratio,
end2end delay, and throughput was compared. For better understanding of results
comparison, we introduce these criteria.
a) Packet delivery ratio- It expresses the ratio of the total number of publication
messages received by each subscriber node, up to the total number of
publication messages generated by all publisher nodes of the events to which
the subscriber node has subscribed.

It can be calculated by the following formula:

PDR= ((total packets-loss)/total packets)*100

b) End2End Delay- The delay of a packet in a network is the time it takes the
packet to reach the destination after it leaves the source.

c) Throughput – Throughput is the number of packet that is passing through the

channel in a particular a unit of time. This performance metric show the total
number of packets that have been successfully delivered from source node to
destination node and it can be improved with increasing node density.

The amount of samples generated by the network as response to a given query is

equal to the number of sensors, k, that are present and active when the query is
received.
It can be calculated by the following formula:
Throughput=total packets/End2EndDelay

37 | P a g e
 Table 4.1:Different Parameters used for Feed Forward Back Propagation

Parameters for Packet Packe Packet End2End Throughpu

Feed-Forward transmissio t Delivery Delay(e2edela t
Backpropagatio n Drop Ratio(pdr y)
n )

200 23.10 88.45 0.1272 1.5726

200 16.80 91.60 0.1256 1.5921
200 33.60 83.20 0.1730 1.1562
200 16.80 91.60 0.1335 1.4976
200 21.00 89.50 0.1331 1.5022
200 25.00 87.40 0.1667 1.1994
210 16.08 92.00 0.1236 1.6993
210 29.40 86.00 0.1745 1.2036

Table 4.2: Different Parameters used for Hopfield neural network

Parameters Packet Packet Packet End 2End Throughput

for transmission Drop Delivery Delay(e2edelay)
Hopfield Ratio(pdr)
NN
200 10.50 94.75 0.0868 2.3034
200 10.50 94.75 0.0883 2.2656
200 10.50 94.75 0.0893 2.2391
200 10.50 92.65 0.1280 1.5629
200 10.50 94.75 0.0876 2.2824
210 12.60 94.00 0.0895 2.3459
210 14.70 92.65 0.1280 1.5629

38 | P a g e
4.1 SNAPSHOTS/ GRAPHS:
4.1.1 SNAPSHOTS:

This figure shows the data /packet transmission from one node to another. Here the
node which send data from one node to another , all are static.

Collided/dead
node

Figure 4.1: Matlab simulation for Data/Packet Transmission.

39 | P a g e
This figure shows a sensor node (yellow) which is in rectangle shape. When more
than one node sends the data simultaneously at the same node then the sensor node
sense it and change the color from green to yellow.

Sensor Node

Figure 4.2 : Matlab Simulation of a Sensor Node.

This figure shows a collided/dead node .When more than one node sends the data to
same node at the same time, then there is chances of collision. Here as the collision
occur at another rectangle node; it will change his color from green to red.

40 | P a g e
 Collided
Node/Dead
Node

Figure 4.3 : Matlab simulation of a Collided/Dead Node.

41 | P a g e
This figure shows that after using Hopfield neural network, we can avoid the
collision.

A Neural
Network
Approach

Figure 4.4: Matlab simulation for Neural Network Approach.

This figure shows collision avoidance after using Hopfield neural network. Here the
two rectangle node shows that there is no collision when two or more node sends the
data at the same time and there is no changes in node’s color.

42 | P a g e
 Collision avoidance
using Hopfield NN

Figure 4.5: Matlab simulation for collision avoidance.

43 | P a g e
4.1.2 GRAPHS:

Packet Delivery Ratio

96

94

92

90

88
Back Propogation
86 Hop Field
84

82

80

78

76
200 200 200 200 200 200 210
Number of Packet Transmit

Figure 4.6: Graph showing Packet Delivery ratio over no. of Packet Transmit

End2End Delay
Throughput
0.2
2.5
0.18
0.16
0.142
0.12
Back Propagation
0.1
1.5 Hop Field
0.08 Back propagation
0.06 Hop Field
0.041
0.02
0
0.5 200 200 200 200 200 200 210
No. of Packet Transmit

0
200 200 200 200 200 200 210
Figure 4.7: Graph for End2End Delay over no. of Packet Transmit
No. of Packet Transmit

Figure 4.8 :Graph for Throughput over No. of Packet Transmit

44 | P a g e
 CHAPTER: 5
CONCLUSION AND FUTURE WORK

Wireless sensor networks (WSNs) are innovative large-scale wireless networks that
consist of distributed, autonomous, low-power, low-cost, small-size devices using
sensors to cooperatively collect information through infra-structure-less ad-hoc
wireless network. The development of wireless sensor networks was originally
motivated by military applications such as battlefield surveillance. However, wireless
sensor networks are now used in many civilian application areas, including
environment and habitat monitoring, healthcare applications, home automation, and
traffic control.

Security plays a fundamental role in many wireless sensor network applications.

Because sensor networks pose unique challenges, security techniques used in
conventional networks cannot be directly applied to WSNs because of its unique
characteristics. First, sensor nodes are very sensitive of production cost since sensor
networks consist of a large number of sensor nodes argued that the cost of a sensor
node should be much less than one dollar in order for sensor networks to be feasible.
Therefore, most sensor nodes are resource restrained in terms of energy, memory,
computation, and communication capabilities. Normally sensor nodes are powered by
batteries, and recharging batteries are infeasible in many circumstances. Energy
consumption becomes akey consideration for most sensor network protocols. Second,
Sensor nodes may be deployed in public hostile locations, which make sensor nodes
vulnerable to physical attacks by adversaries. Generally, adversaries are assumed to
be able to undetectably take control of a sensor node and extract all secret data in the
node.

Furthermore, the scale of sensor networks is considerably large, and the network
topology is dynamically adjusted, because some nodes may die out of running out of

45 | P a g e
energy or failure, and new nodes may join the network to maintain desirable
functionality. At last, sensor networks use insecure wireless communication channel
and lack infrastructure. As a result, existing security mechanisms are inadequate, and
new approaches are desired. Since large number of sensor nodes are densely
deployed, neighbor nodes may be very close to each other. Hence, multihop
communication in sensor networks is expected to consume less power than the
traditional single hop communication. Furthermore, the transmission power levels can
be kept low, which is highly desired in covert operations.

Multihop communication can also effectively overcome some of the signal

propagation effects experienced in long-distance wireless communication. One of the
most important constraints on sensor nodes is the low power consumption
requirement. Sensor nodes carry limited, generally irreplaceable, power sources.
Therefore, while traditional networks aim to achieve high quality of service (QoS)
provisions, sensor network protocols must focus primarily on power conservation.
They must have inbuilt trade-off mechanisms that give the end user the option of
prolonging network lifetime at the cost of lower throughput or higher transmission
delay. Many researchers are currently engaged in developing schemes that fulfill these
requirements.

The main feature of WSN that makes it unique is its flexibility in terms of the shape
of the network and mobility of the sensors. In this thesis, we have proposed the
throughput, delay and packet delivery ratio of networks according to no. of packet
transmit using hop-field neural network. We have compared the results of back-
propagation technique with Hop-field neural network. The comparison shows that
results of hop-field are better than the back propagation. Here, the packet delivery
ratio and through-put are increased and end-to-end delay is decreased. This thesis also
explores solutions to efficiently recover collisions in WSNs. In future neuro-fuzzy and
fuzzy applications can be apply to avoid the collision using these parameters.

46 | P a g e
 References

[1] C. Karlof and D. Wagner, “Secure routing in wireless sensor networks: Attacks
and countermeasures”Ad Hoc Networks, vol. 1, no. 2, 2003
[2] V. Arnaudov, “Unified Management of Heterogeneous Sensor Networks In the
Atlantis Framework”, Department of Computer Science, Brown University.
[3] M. Sims, C. Goldman, V. Lesser, “Self-Organization through Bottom-up coalition
formation”, University of Massachusetts.
[4] J. Suh, Mike Horton, “Powering sensor networks”, Potentials, IEEE,
August/September 2004.

[5] K. L. Chee, P. K. Sivaprasad, S.V. Rao, J.G. Lim, “Clock Drift Reduction For
Relative Time Slot TDMABased Sensor Networks”, Personal Indoor and Mobile
Radio Communications, PIMRC 2004. 15th IEEE International Symposium, Pages:
1042 – 1047, Vol.2, 5-8 Sept. 2004.

[6] M. Hempstead, N. Tripathi, P. Mauro, G. Y. Wei, David Brooks, “An Ultra Low
Power System Architecture for Sensor Network Applications”, Intelligent Sensors,
Sensor Networks and Information Processing Conference, Proceedings of the 2004
14-17 Dec. 2004, Pages: 13 – 18, 2004.

[7] S. K. Jayaweera, “An Energy-efficient Virtual MIMO Communications

Architecture Based on V-BLAST Processing for Distributed Wireless Sensor
Networks”, Sensor and Ad Hoc Communications and Networks, 2004. IEEE SECON
2004. First Annual IEEE Communications Society Conference, Pages: 299 – 308,
October 2004

[8] I. Akyildiz, W. Su, Y. Sankarasubramanian, E. Cayirci, “A Survey on Sensor

Networks”, IEEE Communications Magazines, August 2002.

47 | P a g e
[9] SinchanRoychowdhury, ChiranjibPatra,” Geographic Adaptive Fidelity and
Geographic Energy Aware Routing in Ad Hoc Routing”, Special Issue of IJCCT
Vol.1 Issue 2, 3, 4; 2010 for International Conference [ACCTA-2010], 3-5 August
2010.
[10] TahirNaeem, Kok-Keong Loo, “Common Security Issues and Challenges in
Wireless Sensor Networks” and IEEE 802.11 Wireless Mesh Networks, International
Journal of Digital Content Technology and its Applications, Page 89-90 Volume 3,
Number 1, year 2009
[11] John Paul Walters, Zhengqiang Liang, Weisong Shi, VipinChaudhary, “Wireless
Sensor Network Security: A Survey”, Security in Distributed, Grid and Pervasive
Computing Yang Xiao (Eds), Page3-5, 10-15, year 2006

[12] Pfleeger, C. P. and Pfleeger, S. L., “Security in Computing”, 3rd edition, Prentice
Hall 2003
[13] Culpepper, B.J. and Tseng, H.C., “Sinkhole intrusion indicators in DSR
MANETs”, Proc. First International Conference on Broad band Networks, 2004, pp.
681 – 688.
[14] C. Karlof and D. Wagner, “Secure routing in wireless sensor networks: Attacks
and countermeasures,” Elsevier’s AdHoc Networks Journal, Special Issueon Sensor
Network Applications and Protocols, vol. 1, no. 2–3, pp. 293– 315, September 2003.

[15] Karlof, C. and Wagner, D., “Secure routing in wireless sensor networks: Attacks
and countermeasures”, Elsevier's Ad Hoc Network Journal, Special Issue on Sensor
Network Applications and Protocols, September 2003, pp. 293-315.
[16] Y. Hu, A. Perrig, and D. Johnson, “Wormhole detection in wireless ad hoc
networks,” 2002.[Online]. Available: citeseer. ist.psu.edu/hu02wormhole.html
[17] Mauro Conti,Roberto Di Pietro,Luigi Vincenzo Mancini, and Alessandro
Mei,“Distributed Detection of Clone Attacks in Wireless Sensor Networks” IEEE
Transaction on dependable & secure computing vol.8 n0.5 September 2011.
[18] Blackert, W.J., Gregg, D.M., Castner, A.K., Kyle, E.M., Hom, R.L., and Jokerst,
R.M., “Analyzing interaction between distributed denial of service attacks and
mitigation technologies”, Proc. DARPA Information Survivability Conference and

48 | P a g e
Exposition, Volume 1, 22-24 April, 2003, pp. 26 – 36.
[19] Wang, B-T. and Schulzrinne, H., “An IP traceback mechanism for reflective
DoS attacks”, Canadian Conference on Electrical and Computer Engineering,
Volume 2, 2-5 May 2004, pp. 901 – 904.

[20]Q. Ren and Q. Liang, "Secure media access control (MAC) in wireless sensor
networks: Intrusion detections and countermeasures," in Proc.15th IEEE Int.
Symposium on Personal, Indoor and Mobile RadioCommunications PIMRC 2004,
vol. 4, 5-8 Sept. 2004, pp. 3025-3029.

[21] "Fuzzy logic-optimized secure media access control (FSMAC) protocol wireless
sensor networks," in Proc. IEEE Int. Conf. on ComputationalIntelligence for
Homeland Security and Personal Safety CIHSPS2005, March 31 2005-April 1 2005,
pp. 37-43.

[22]Q. Ren and Q. Liang, "Secure media access control (MAC) in wireless sensor
networks: Intrusion detections and countermeasures," in Proc.15th IEEE Int.
Symposium on Personal, Indoor and Mobile RadioCommunications PIMRC 2004,
vol. 4, 5-8 Sept. 2004, pp. 3025-3029.

[23]"Fuzzy logic-optimized secure media access control (FSMAC) protocol wireless

sensor networks," in Proc. IEEE Int. Conf. on ComputationalIntelligence for
Homeland Security and Personal Safety CIHSPS2005, March 31 2005-April 1 2005,
pp. 37-43.

[24] RishavDubey, Vikram Jain, Rohit Singh Thakur,

SiddharthDuttChoubey,“Attacks in Wireless Sensor Networks” ,International Journal
of Scientific & Engineering Research, Volume 3, Issue 3, March-2012- ISSN 2229-
5518.

[25] Rajkumar, Sunitha K.R and Dr. H.G Chandrakanth “A Survey on Security
Attacks in Wireless Sensor Network”, International Journal of Engineering Research
and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-

49 | P a g e
August 2012, pp.1684-1691.

[26] WazirZada Khan Yang Xiang Mohammed Y Aalsalem, “Comprehensive Study

of Selective Forwarding Attack in Wireless Sensor Networks”, I.J. Computer Network
and Information Security, 2011, 1, 1-10 Published Online February 2011 in MECS
(http://www.mecs-press.org/).

[27] Chaudhari H.C. and Kadam L.U. “ Wireless Sensor Networks: Security, Attacks
and Challenges”. International Journal of Networking Volume 1, Issue 1, 2011, pp-
04-16 Available online at: http://www.bioinfo.in/contents.php?id=108.

[28]JiyongSon ; Dept. of Electr. Eng., Korea Univ., Seoul, South Korea ; Hwan-
JooKwak ; Gwi-Tae Park ,member IEEE proposed some work” Back propagation
neural network based real-time self-collision detection method”2011.

[29] Kalpana Sharma, M K Ghose, “Wireless Sensor Networks: An Overview on its

Security Threats”, IJCA Special Issue on “Mobile Ad-hoc Networks” MANETs,
2010.

[30] Dr. G. Padmavathi and Mrs. D. Shanmugapriya, “A Survey of Attacks, Security

Mechanisms and Challenges in Wireless Sensor Networks”, (IJCSIS) International
Journal of Computer Science and Information Security, Vol. 4, No. 1 & 2, 2009

[31]Ahmad Hosseingholizadeh Dr. AbdolrezaAbhari Department of Computer

Science Ryerson University Toronto, Canada proposed some work “A neural network
approach for Wireless sensor network power management”2009.

[32] Hong-Ning Dai proposed some work” Throughput and Delay in Wireless Sensor
Networks”2009.

[33] Raghavendra V. Kulkarni, “Neural Network Based Secure Media Access Control
Protocol for Wireless Sensor Networks” Senior Member, IEEE,” 2009.

50 | P a g e
[34] David Boyle, “Securing Wireless Sensor Networks: Security Architectures”
JOURNAL OF NETWORKS, VOL. 3, NO. 1, JANUARY 2008.

[35] Phillip Reindl and Kendall Nygard , “Defending Malicious Collision Attacks in
Wireless Sensor Networks”. Department of Computer Science North Dakota State
University Fargo,2007.

[36] Frank Oldewurtel, “Neural Wireless Sensor Networks”2005.

[37] Kai Xing, ShyaamSundhar and RajamadamSrinivasan,“Attacks and

Countermeasures in Sensor Networks: A Survey”, NETWORK SECURITY Scott
Huang, David MacCallum, and Ding Zhu Du(Eds.) pp.@2005 Springer.

[38] K. Jamieson, H. Balakrishnan, proposed some work “A mac protocol for event-
driven wireless sensor networks”2003.

[39] Simon X. Yang, Member, IEEE,proposed some work “Neural Network

Approaches to Dynamic Collision-Free Trajectory Generation”2001.

[40] J. Polastre, J. Hill, and D. Culler, “Versatile Low Power Median Access for
Wireless Sensor Networks,” in ACM SenSys, 2004.

[41]Hopfield, J. J., Brody, C. D. and Roweis, S. (1998) Computing with action

potentials. Proceedings of Advances in Neural Information Processing Systems10,
Denver, CO, 1-6 December, pp. 166-172. MIT Press, Cambridge, MA.2.
[42] H. He, Z. Zhu, “A Neural Network Model to Minimize the Connected
Dominating Set for Self-Configuration of Wireless Sensor Networks,” IEEE
Transaction on neural networks, vol. 20, No. 6, pp. 973-982, 2009.

[43]Hopfield, J. J., Brody, C. D. and Roweis, S. (1998) “Computing with action

potentials. Proceedings of Advances in Neural Information Processing Systems”10,
Denver, CO, 1-6 December, pp. 166-172. MIT Press, Cambridge, MA.2.

51 | P a g e
[44] Simon X. Yang, Member, IEEE,proposed some work “Neural Network
Approaches to Dynamic Collision-Free Trajectory Generation”2001.

[45],K. Jamieson, H. Balakrishnan, proposed some work “A mac protocol for event-
driven wireless sensor networks”2003.

[46] L. Fausett, Fundamentals of Neural Networks. Prentice Hall, 1994.

[47] D. J. C. MacKay, Information Theory, Inference, and Learning Algorithms.

Cambridge University Press, 2003.

[48] J. J. Hopfield, “Neural networks and physical systems with emergent collective
computational abilities,” Proc. of National Academy of Sciences,
1982.

[49] The MathWorks Inc. MATLAB 7.0 (R14SP2). The MathWorks Inc., 2005.
[50] S. J. Chapman. MATLAB Programming for Engineers, Thomson, 2004
[51] http://en.wikipedia.org/wiki/MATLAB
[52] http://wiki.answers.com/Q/What_are_the_applications_or_uses_of_matlab
[53] A. Gilat. MATLAB: An introduction with Applications. John Wiley and Sons,
2004.
[54] J. Cooper. A MATLAB Companion for Multivariable Calculus.Academic Press,
2001.

[55] Ian F. Akykildiz, Weilian Su, YogeshSankarasubramaniam, and ErdalCayirci, “A

Survey on Sensor Networks”, IEEE Communication Magazine, year 2002.

[56] John Paul Walters, Zhengqiang Liang, Weisong Shi, VipinChaudhary, “Wireless
Sensor Network Security: A Survey”, Security in Distributed, Grid and Pervasive
Computing Yang Xiao (Eds), Page3-5, 10-15, year 2006.

[57] Y. Wang, G. Attebury, and B. Ramamurthy, “A Survey of Security Issues in

Wireless Sensor Networks,” IEEE Commun.SurveysTutorials, vol. 8, pp. 2–23, year

52 | P a g e
2006.
[58] FengZhao,LeonidasGuibas,,”Wireless Sensor Networks”, Morgan Kaufmann
Publications.

53 | P a g e