Various Papers in Our College

Proceedings of the Seventh NCVIT
7
th
March 2014 GCT,Coimbatore

To Enhance the Lifetime of WSN
Network using PSO

K.SyedAliFathima, T.Sumitha,
Assistant Professor, PG Student,
M. Kumarasamy College of engineering, M.Kumarasamy College of engineering,
Karur. Karur.
sumithamecse@gnail.com

.
Abstract The major issues in wireless sensor networks is
to maximize the network lifetime of sensors.The best
energy efficient protocol is LEACH to reduce the energy
consumption and it can extend the lifetime of wireless
sensor network.Clustering techniques can be used to
communicate with cluster-head and base station. If the
base station is far away from the cluster-head ,energy
consumption will be increased and it can reduce the
lifetime of wireless sensor network.To overcome these,
Paticle swarm Optimization technique is implemented
with this protocol inorder to achieve maximum lifetime of
wireless sensor network.PSO is used to extend the scalable
and energy efficiency. It is easy to implement and the
mutation calculation speed is very fast.

Keywords Low Energy Adaptive Clustering
Hierarchy (LEACH) , Wireless Sensor Networks (WSNs) ,
Partical Swarm Optimization(PSO) ,Energy efficiency
and Cluster methods .

I.INTRODUCTION
A wireless sensor network consists of a large number
of sensor nodes and it can be used as an effective tool for
collecting data from various situations. The major issue in
wireless sensor networks is developing an energy-efficient
routing protocol which has a significant impact on the overall
lifetime of the sensor network.
The wireless sensor networks consist of more
number of nodes and that nodes are transfer the information
from source to sink or base station. The components of sensor
networks are sensor nodes, sink, base station and sensor field.

Fig 1: Wireless sensor networks

Network routing protocols are in charge of routing scheme as
well as maintaining the network structure in WSNs. There are
three types of network structure: flat routing, hierarchical
routing and location-based routing [4]. In wireless sensor
network, LEACH protocol is one of the best energy efficient
protocol. It helps to reduce the energy dissipation and it is a

first hierarchical based network routing protocol. Hierarchical
routing is mainly considered as two layer architecture.
Therefore, one layer is engaged in cluster head selection and
the other layer is responsible for routing.
LEACH uses a clustering method to reduce the
energy consumption and it arranges the nodes in network as
small clusters and it select one as cluster head (CH). It
provides the balancing of energy usage by random rotation of
cluster heads. LEACH protocol uses a data fusion algorithm
for reduce the data transmission. Cluster head are used to
compress and reduce the information where received from all
nodes and sends it to the sink.

LEACH operations are divided into two phases:

1. Setup phase
2. Steady phase

7
th

Fig 2: Clustering methods

In setup phase , each node is independent of other
node. Then it will form cluster and cluster head(CH) is
chosen for each cluster. Setup phase are used to minimize the
overhead cost. During steady phase, the sensor nodes the non-
cluster head nodes starts sensing data and sends it to their
cluster-head according to the TDM schedule. The cluster-head
node, after receiving data from all the member nodes,
aggregates it and then sends it to the base-station . Steady
Phase consists of Schedule Creation and Data Transmission.
LEACH protocol periodically elects the cluster head nodes
and re-establishes the clusters according to a round time,
which ensures energy dissipation of each node in the network
is relatively evenly [3].

II.OVERVIEW
In this section, the LEACH protocol and PSO
techniques are explained.

LEACH PROTOCOL:

Wireless sensor networks is used for developing a
routing protocol, has a significant impact on overall lifetime
of the sensor network which employs a new technique of
LEACH protocol called VLEACH.The central role is to
reduce energy consumption in sensor network.LEACH
performs self-organizing and re-clustering functions for every
round [2][8]. Sensor nodes organize themselves into clusters
in LEACH routing protocol. LEACH-E proposed to elect the
cluster-heads according to the energy left in each node.[11] In
every cluster one of the sensor node acts as cluster-head and
remaining sensor nodes as member nodes of that cluster. Only
Cluster-head can directly communicate to sink and member
nodes use cluster-head as intermediate router in case of
communication to sink.
The potential problem in current protocols is that
they find the lowest energy route and use that for every
communication. We propose a new protocol that we call
energy aware routing. This is used to increase the survivability
of networks. Additionally, these sensor nodes have limited
processing power, storage and energy, while the sink nodes
have powerful resources to perform any tasks or communicate
with the sensor nodes.. Then we propose a heuristic routing
algorithm to achieve our design goal. The algorithm works in
the following way. First, we compute the network throughput,
which is the most important performance metric for data-
intensive computations, according to the routing on all data
centre switches. The corresponding routing is called basic
routing. Second, we gradually remove switches from the basic
routing, until when the network throughput decreases to a
predefined performance threshold. Third, switches not
involved in the final routing are powered off or put into sleep
mode. However, to save energy, sensor nodes send their
messages to their CHs, which then aggregate the messages,
and send the aggregate to the BS. However, because it is a
cluster based protocol, relying fundamentally on the CHs for
data aggregation and routing, attacks involving CHs are the
most damaging. If an intruder manages to become a CH, it can
stage attacks such as sinkhole and selective forwarding, thus
disrupting the workings of the network. To overcome the
disadvantages of LEACH protocol, PSO technique is
employed.

PARTICLE SWARM OPTIMIZATION (PSO):

Particle swarm optimization (PSO) is a simple,
effective and efficient optimization algorithm. PSO is used to
explore the search place. It is easy to implement and it can be
applied for both scientific research and engineering use.
In PSO, a global fitness function is used by all the
particles in the swarm. In this, No overlapping and mutation
calculation speed is very fast. It evaluates the fitness of each
particle. It occupies the bigger optimization ability and it
complete easily. Particles in traditional PSO represent the
candidate solutions to a single optimization problem. [5]. the
energy consumption and reliability are taken into an account
topology control is the problem of LEACH protocol. The
binary particle swarm optimization (BPSO) approach to solve
the disjoint set covers (DSC) problem in the wireless sensor
networks (WSN). The DSC problem is to divide the sensor
nodes into different disjoint sets and schedule them to work
one by one in order to save energy while at the same time
meets the surveillance requirement, e.g., the full coverage
objective of DSC is to maximal the number of disjoint. PSO
based algorithm is used to locate the optimal sink position to
the nodes to make the network is more energy efficient.
Some of the techniques are used to improve the
network lifetime of wireless sensor network:
Data fusion algorithm.
Energy-efficient routing.
7
th

Locating optimal sink position.
PSO is more robust and easy to reach the solution for real
world environmental monitoring and data aggregation
problems.

Fig: Particle swarm optimization.

III. ENERGY ANALYSIS

Energy analysis of routing protocol are as follows:
Sensor sends the data directly to sink. If the sink is far away,
large amount of transmit power from each note will quickly
drain nodes and it reduce system lifetime. By using routing
protocol each node is acts as routers for other node's data in
addition to sensing data. And this protocol is used to neglect
energy dissipation of receiver intermediate nodes. While
creating the infrastructure, the process of setting up the routes
is greatly influenced by energy considerations. The multihop
routing will consume less energy. So multi-hop routing
introduces significant overhead for topology management and
medium access control. In sometime sensors are scattered
randomly over an area and multi-hop routing becomes
unavoidable [13] for large area energy consumptions,
Improved FZ-LEACH has been introduced.

A.IMPROVED FZ-LEACH:

For large scale deployments, very small clusters does
not provide energy efficiently. So, it decreases the network
lifetime of wireless sensor network. The new energy efficient
clustering protocol is Improved FZ-LEACH. It can eliminate
the Far-Zone problem. Far-Zone is a group of sensor nodes
which are placed at locations where their energies are lesser.
The communication between the nodes and Sink is based on
the energy consumption . The communicating nodes are in
active mode and the remaining nodes will be in sleep mode,
for this sleep scheduling algorithm has been used. LEACH-C
each node sends their current location information and
residual energy level to the sink. The Improved FZ-LEACH
algorithm outperforms LEACH in terms of energy
consumption and network lifetime. [12]

Energy efficiency is essential in some applications of
wireless sensor network, especially when sensor nodes are
situated in non-accessible areas like battlefield [9]. For such
kind of applications solar-ware LEACH (sLEACH) has been
included to maximize the lifetime of wireless sensor network
through solar power. sLEACH some nodes are facilitated by
solar power and these nodes will act as cluster-heads mainly
depending upon their solar status. Both LEACH and LEACH-
C are extended by sLEACH [6].

SOLAR-AWARECENTRALIZEDLEACH:

By using Central control algorithm, solar-aware
Centralized LEACH cluster head are selected by Base station.
Base station normally select solar powered nodes which have
maximum residual energy. In sLEACH nodes transmit their
solar status to base station along with energy and nodes with
higher energy are selected as cluster-head. Performance of
sensor network is increased when number of solar-aware
nodes is increased. Sensor network lifetime also depends upon
the sunDuration. If the sunDuration is smaller cluster-head
handover is also performed in sLEACH [9].
If node serving as cluster- head is running on battery
and a node in cluster send data with flag, it denoted as solar
power is increased, and this node will become cluster-head
instead of place first its serving as cluster-head.

IV. CLUSTERING IN HIERRACHIAL ROUTING

LEACH protocol is a si ngl e-hop cl ustering
rout i ng protocol i n wi r el ess sensor net wor k.
Cl ust er s i n LEACH are for med dynami cal l y and
per iodical l y, whi ch changes i nt er act ions among t he
nodes and requir es t hat any node needs to be ready
t o j oi n wi t h any CH at any t i me[9].

7
th

Fig: LEACH clustering communication hierarchy for WSNs.

All nodes in the cluster transmit their data packets to
CHs in specified TDMA time. TDMA-based protocols are
naturally energy preserving, because time slots are built-in,
and it does not suffer from collisions. The CH for each
cluster receives and aggregates the data from cluster and then
transmits the aggregated data to the BS. The other clustering
algorithm are proposed based on LEACH, such as TEEN
(Threshold Sensitive Energy Efficient Sensor Network
Protocol)[4] , PEGASIS(Power Efficient Gathering in Sensor
Information Systems)[5] HEED(Hybrid Energy-Efficient
Distributed Clustering)[6] [10].

A.TWO LEVEL-LOW ENERGY ADAPTIVE
CLUSTERING HIERARCHY(TL-LEACH)

When compared to LEACH protocol, build a two-
level hierarchy protocol to save better energy consumption.
The TL-LEACH uses random rotation of local
cluster base stations (primary cluster-heads and secondary
cluster-heads). [14]It uses data fusion concepts to enable a
scalability and robustness. It is used to reduce the energy
consumption. The primary cluster-head are used to creates a
TDMA schedule assigning each node a slot to transmit. Each
primary cluster-head chooses a CDMA code and informs all
the nodes at second level in its group to transmit the
information.

DRAWBACKS:

No Guaranteeing that the cluster head nodes are
distributed through the network.
Time duration of the set-up phase is non-
deterministic and the collisions will cause the time
duration too long and the sensing services are
interrupted.
LEACH uses dynamic clustering which results in
extra overhead.
Cluster heads used in the LEACH will consume a
large amount of energy.

V.CONCLUSION AND FUTUREWORK

In this paper, LEACH protocol is used for energy efficiency in
wireless sensor network .Clustering technique is used in
LEACH protocol to improve the lifetime of network .It will
increase the energy but still it need some energy. In future
work include Particle swarm optimization technique with
LEACH protocol it provides more energy efficient to increase
network lifetime of wireless sensor network.

References

[1]J.Gnanambigai, Dr.N.Rengarajan, K.Anbukkarasi,
Leach and Its Descendant Protocols: A Survey
International Journal of Communication and Computer
Technologies Volume 01 No.3, Issue: 02 September
2012.

[2]Yun Li1,2, Nan Yu1, Weiyi Zhang 3, Weiliang Zhao
1, Xiaohu You2, Mahmoud Daneshmand,Enhancing the
Performance of LEACH Protocol inWireless Sensor
Networks,IEEE INFOCOM 2011 Workshop on
M2MCN-2011.

[3]Ankita Joshi 1, Lakshmi Priya.M 2 A Survey of
Routing Protocols for Wireless Sensor Networks .

[4]Raghavendra V. Kulkarni, Senior Member, IEEE, and
Ganesh Kumar Venayagamoorthy, Senior Member,
IEEE Particle Swarm Optimization in Wireless Sensor
Networks: A Brief Survey.

[5]Jenn-Long Liu and Chinya V. Ravishankar, Member,
IEEE LEACH-GA: Genetic Algorithm-Based Energy-
Efficient Adaptive Clustering Protocol for Wireless
Sensor Networks, International Journal of Machine
Learning and Computing, Vol.1, No. 1, April 2011.

[6]Qian Liao, Hao Zhu An Energy Balanced
Clustering Algorithm Based on LEACH Protocol,
Proceedings of the 2nd International Conference On
Systems Engineering and Modeling (ICSEM-13).

[7]Nutan Sindhwani1, Rohit Vaid2 V LEACH: AN
ENERGY EFFICIENT COMMUNICATION
PROTOCOL FOR WSN, Nutan Sindhwani, Rohit Vaid
/ Mechanica Confab ISSN: 2320-2491.

[8]Md. Junayed Islam, Md. Muhidul Islam, Md. Nazrul
Islam A-sLEACH : An Advanced Solar Aware Leach
Protocol for Energy Efficient Routing in Wireless
Sensor Networks. Proceedings of the Sixth
International Conference on Networking (ICN'07).

[9]Vinay Kumar1, Sanjeev Jain2 and Sudarshan Tiwari,
IEEE Member3 Energy Efficient Clustering Algorithms
in Wireless Sensor Networks: A Survey, IJCSI
International Journal of Computer Science Issues, Vol.
8, Issue 5, No 2, September 2011.

[10]Sonu Vashist1, Jitender Khurana2 Enhanced
LEACH Protocol for Wireless Sensor Networks. ISSN
7
th

2250-2459, ISO 9001:2008 Certified Journal, Volume 3,
Issue 6, June 2013.

[11]Ms.V.MuthuLakshmi Advanced LEACH Protocol
in Large scale Wireless Sensor Networks, International
Journal of Scientific & Engineering Research, Volume
4, Issue 5, May-2013 .

[12]Kemal Akkaya and Mohamed Younis A Survey on
Routing Protocols for Wireless Sensor Networks.

[13] V. Loscr, G. Morabito, S. Marano A Two-Levels
Hierarchy for Low-Energy Adaptive Clustering
Hierarchy (TL-LEACH).

7
th

Real time data monitoring in smart transmission grid using wireless sensors

K.Venkatasubramani, PG Scholar
Department of Electrical and Electronics Engineering
M.Kumarasamy college of Engineering
Karur-639113, Tamil Nadu, India
venkatakalidass@gmail.com

R.Karthikeyan, Professor
Department of Electrical and Electronics Engineering
M.Kumarasamy college of Engineering
Karur-639113, Tamil Nadu, India
papkarthik@gmail.com
AbstractIn this paper, the design of three stage hybrid
architecture for controlling and preventing certain
disturbances caused in transmission lines is studied. The
transmission lines of modern power systems are equipped
with Wireless Sensor Network (WSN). Since WSNs are
capable of cost efficient monitoring over a wide range of
geographical areas. Consequently monitoring of mechanical
parameters of transmission line in smart grid is achieved.
The hybrid architecture composed of three stages wired,
wireless and cellular technologies. The main intention of
this paper is to study the cost efficient monitoring of various
mechanical parameters which affecting the transmission
line in smart grid. We solved the placement problem for the
optimal placement of cellular towers to minimize the
installation and operational costs.
KeywordsHybrid architecture; placement problem;
wireless sensor networks; transmission lines; operational
costs
I. INTRODUCTION
Power system operators need to operate the transmission
systems under complex situations and atmosphere. So the
current monitoring, analysis and control strategy for
transmission networks may not be able to meet increasingly
diverse challenges. Most of the transmission lines currently
used is highly vulnerable to many forms of natural and
manmade disaster, which can hardly affects the efficiency and
stability of the grid. Hence by replacing the age old
transmission lines with good communication network, the
transmission process can be improved. Wireless sensor based
communication networks solves the following concerns.
Concerns like real time structural framework, accurate fault
diagnosis by identification and differentiation of electrical
faults from the mechanical faults, cost reduction, maintenance.
The use of sensor network dealt with several other applications
like mechanical state processing and dynamic transmission line
ratings. By using wireless networks we can achieve faster
delivery of enormous amount of highly reliable information.
The proposed network is able to transport sensitive data such as
current state of the transmission line and its control to and from
smart grid. Our main objective in this paper is to design a
communication framework to transport enormous data in low
costs. By using Supervisory Control and Data Acquisition
System (SCADA) in substations, the enormous amount of
sensitive data can be communicated easily with faster response.
In transmission lines, if any disturbance occurs it is difficult to
repair other than substations and distribution stations. Due to its
large geographical coverage area the task of locating the faulted
area is very much difficult. The recent blackouts in U.S. and
Northern India have shown that the failure to access and
understand the condition of the power system. And delay in
taking appropriate corrective actions after an outage can lead to
widespread blackout of large areas of power system. Hence
smart grids are equipped with extra communication networks to
solve the above concerns. Thus the features of the smart gird
are discussed in order to provide better results. The smart grid
represents the full suite of current and proposed responses to the
challenges of power supply. Because of the miscellaneous
range of factors there are numerous competing taxonomies and
no accord on a universal description. The features of the smart
power grid are listed below.
- Reliability
- Flexibility
- Efficiency
- Load adjustment
- Peak curtailment
- Sustainability
- Bidirectional energy flow
We propose the use of wireless sensor network
technology for detection of mechanical disturbances in
transmission lines such as: conductor failure, tower collapses,
hot spots, wind conditions, etc. The proposed design involves
the installation of sensors for mechanical monitoring in
predetermined towers of a transmission lines and communicate
via wireless networks. The main goal is to obtain a complete
physical and electrical model of the power system in real time,
diagnose permanent as well as temporary faults and to make
security for extreme mechanical conditions And also placement
of cellular towers in the optimal location is done so as to
enhance low extreme installation and operational costs.
II. RELATED WORK
Several works and proposals have been made to
improve the state of art in deployment of multiple wireless
sensors to monitor the various mechanical parameters. In this
work the goal is to install the reliable wireless sensors in
particular vulnerable location of the transmission lines. So the
sensed data should be transmitted to the control centre via
proper communication wireless networks [1]. Due to the vast
geographical expanse of transmission line infrastructure,
wireless networking provides a feasible and cost effective for
7
th

monitoring of long distance transmission line. In the previous
papers we study, authors develop a quadratic equation based
solution for finding the optimal locations of cellular
transceivers the objective is to minimize the delay in
information delivery [2]. We distinct this work on the following
grounds:
- The framework formulated and presented by the
authors in reference paper relies strongly on proportion
[3]. The core network infrastructure and the cellular
infrastructure are implicit to be symmetric and
accessible at all periods. As well, it is implicit that all
transmission towers are identical and transfer same
quantity of data. However certain parameters bring in
irregularity as enumerated below:
-Bare Cellular Coverage (due to unavailability of
cellular towers in the region) or cellular outage.
-Variation in the content of data transmitted by the
towers in space of its location or environment.
-Serrated terrain in certain regions of the transmission
line limits the usage of wireless devices and forcing
the use of cellular networks.
- The evaluation done before uses minimizing delay as a
goal. While cost concerns are mentioned, deployment
and protection costs are not used as factors restraining
the number of cellular transceivers.
- The method used already is quadratic equation for
optimal placement of cellular transceivers. Roots of
quadratic equations are rounded off to depict the
number of cellular enabled towers. This leads to
erroneous outcome. Also factors such as latency and
bandwidth affect the placement of cellular
transceivers.
In this paper, we propose an optimal solution which
minimizes installation and operational costs while satisfying all
the constraints such as latency and bandwidth. We provide a
generic presentation for enhancing challenges such as
asymmetric flow bandwidth, irregular cellular coverage, etc.
Further our proposed method also provides a cost optimal
deployment of cellular towers.
III. DESIGN OF WIRELESS NETWORK
For designing a robust wireless sensor networks many
factors such as latency, resiliency, security and bandwidth
constraints are taken into account [4]. While low cost of these
wireless sensors gains large scale installation and less
safeguarding cost. Transmission towers are deployed in a linear
arrangement sharing hundreds of miles. In order to provide
smart communication bandwidth is required to provide intended
data to reach its destination in a given time. While performing
literature survey for our studies, we came to notice that the two
level models are, given for supporting the overhead
transmission line monitoring applications [5]. But including the
topological factors of the transmission lines, the less bandwidth,
less data wireless nodes would fail to transfer huge amount of
data in a multi hop manner. The hierarchical model suggested
offers a very costly solution with the ideas of deploying cellular
transceivers on every tower. And this network can bring
extremely low data transmission, the model is cost ineffective
and it gets huge installation and subscription costs. The main
work is to suggest the problem of finding optimal allocation of
cellular transceivers. Fig. 1 shows the proposed framework of
wireless sensors. While studying we faces large consequences
in enumerating the array of challenges associated with
monitoring the wide area network like transmission grid.
Necessary control and preventive measures have to be made
while the sensors provide the faulted data and the physical
structure has to be cleared immediately in short duration of
time. The linear system topology proves to be a major
challenge for wireless network design with respect to latency
constraints and bandwidth constraints. Performance
evaluation of the linear network model shows that successful
delivery ratio of the packets from the nodes far away from
the substation is found to be much less than that of nodes
near the substation because packets from a farther node have
to travel a longer distance and the rate of collision is higher
[6]. The effective monitoring of a large transmission line
network requires a hybrid communication infrastructure. This
hybrid communications can be a combination o f wired (copper
cable) and wireless (cellular/IEEE802.15.4) standards to
enhance the capability of the overall network to meet newer
requirements based on emerging smart grid applications [7].
In this paper, we formulate a hybrid hierarchical
network design problem that can provide cost effective data
transmission while at the same time respecting the bandwidth,
delay, and connectivity constraints. We formulate a
placement problem to optimize the number and location of
the cellular enabled towers to significantly reduce the
operational and installation costs while respecting all the
constraints. The hybrid structure composed of three levels of
technologies. Thus the architecture is explained briefly in the
following sections for future classifications.
IV. THREE LEVEL HIERARCHICAL NETWORK
We propose a hierarchical three level wireless network model
for time critical applications. Each level is equipped with an
array of sensors and transceivers with varied capabilities such
that together they achieve the necessary behavior. The plan
involves the setting up of a private WSN of low cost, low
data rate links, utilization of the existing SCADA network,
and a wide area network such as cellular network comprised
of expensive but high data rate links. The proposed network
makes use of the existing SCADA links (Optical fiber) for
communication between substations and control center and
strategically utilizes the existing cellular network for data
transmission from certain transmission towers directly to the
control center. A set of wireless sensors on each tower is
installed as part of the private WSN. Fi g. 2 depicts a power
transmission corridor with large number of transmission
towers, and two substations, one at each end of the
transmission line, and a control centre. Each level of the network
7
th

forms a cluster supporting many to one communication from all
the nodes in the cluster to the cluster head. The first level of
the network is responsible for collecting information
regarding the tower. It is composed of sensor nodes installed
in each transmission structure forming a sensor array in
tower (SAT). This SAT consists of an array of sensor
modules such as tension sensors, accelerometers, temperature
sensors, tilt sensors, motion sensors, vision-based sensors, and
infrared sensors, etc.
Fig. 1 Block diagram of sensor network
Each tower is equipped with a more sophisticated relay node
with improved computation and communication capabilities.
Data from each sensor in the SAT is transmitted to the relay
node. The relay node is accountable for compressing the
data received from the SAT and transmitting it to the
advanced level. The second level of the network is
accountable for transmission of data from towers that are far
away from the substations. Consider a segment composed of a
few towers in the middle of the transmission line network.
Data from these towers cannot reach either of the substations
due to limited bandwidth of the intermediate wireless links.
In such cases, enabling one of these towers with Cellular
capability can provide a feasible solution as exposed in Fig.
2. It is to be renowned that it is not required to enable all
towers with cellular technology. The second level is thus
composed of segments of such towers transmitting their
aggregated information to the cellular enabled transmission
tower which acts as the head of their segment. The cellular
enabled tower is a transmission tower equipped with an
additional cellular transceiver along with the relay node. This
cellular transceiver offers an alternative way to deliver the
towers data directly to the control center through a high
bandwidth, low latency cellular network. The third level of
the hierarchical network is composed of a single cluster
consisting of two substations and the cellular towers. The
control center acts as the cluster head. Thus, level 1
operates at each tower; level 2 operates at the granularity of a
group of transmission towers. The dimension of the group will
be dictated by the wireless link bandwidth and the required
end to end latency. Level 3 operates at the level of the whole
network where substations and cellular towers transmit to the
control center. Table I summarizes the characteristics of
various communication standards used in this paper.
V. PLACEMENT PROBLEM FORMULATION
In order symbols for placement problem formulation to
provide cost optimized operation in delay constrained and

Fig. 2 Three level hybrid structure
Bandwidth constraints linear networks, the Strategic
placement of cellular transceivers becomes very crucial.
While the cellular transceivers provide low latency, high
bandwidth links, they are costly to install and the
subscription charges can direct the operational cost of the
system. On the other hand, the wireless Zigbee devices are
relatively inexpensive but provide very low data rates. Thus,
there is a tradeoff between cost and delay. In this section, we
first explain our network model and state the placement
problem. Next, we formulate a mathematical program to find
7
th

the optimal location of the cellular enabled towers. Placement
graph is shown in the figure 3. To make optimal placement of
cellular towers it is necessary to study the following concepts.
1. Network representation
2. Placement problem statement
3. Placement problem formulation
1. Network representation: The transmission line is
represented as a graph, G = (V, E). V gives the set of vertices
and E gives the set of edges in G. the total vertices in the graph
is equal to N+3. The set of communication links which can be
wired (SS, CC), cellular (k, CC) or wireless (k, l), where k, l
N. Each link is given by (cij, Bij). Where cij is the operational
cost and Bij is the total bandwidth of the network
representation.
TABLE I
TECHNOLOGIES USED AND CHARACTERISTICS
Properties Optical Fiber Cellular Wireless
Type of link in the
network
Substation to Control
Centre
Transmission towers to Cellular towers Between towers or Between
tower and substation
Bandwidth 10 Gbps Uplink 75 Mbps, Downlink 100 Mbps 250 kbps
Delay 1sec 50 ms 16 ms
Transmission Range 0 since they are
already exist
100m- 10km+ 10m-1.5km
Installation Cost 1x 5x-20x 2x
Channel Contention No No Yes
Subscription fee No Yes No
7
th

2. Placement problem account: G= (V, E) and a set of N flows
find a feasible path for each flow hence the sum of the cost of
all the paths is minimized. If the minimum path chosen by a
tower node is k N includes the edge of (k, CC), then the
cellular tower is placed on the kth tower.
TABLE II
SYMBOLS FOR PLACEMENT PROBLEM
FORMULATION
Symbol Representation
Bij Bandwidth
D Deadline
lijk Latency kth flow on link (i,j)
bk Flow bandwidth for node k
cij Operational cost
IC Installation cost per cellular tower
Yi Binary variable. Is 1 if tower i is cellular
enabled.
X i,j,k Binary variable. Is 1 if k selects (i,j).
Sij Binary variable. Is 1 if (i,j) used by any flow.
3. Placement problem declaration: The algorithms input is the
transmission line of N transmission towers and latency
constraints, D. Table II shows different symbols for problem
formulation. The placement problem can thus formulated as
Minimize:
( , , ) f Si j Yi = t
( , )
,
i j E
cijSi j
e

+
1
.
N
i
ICYi
=

(1)

Subject to :
( , )
, , , ,
i j E
li j kXi j k D
e
s
k N e (2)
( , )
, , 1
i j E
Xi j i
e
=
i N e (3)
\
1 1
, ,
N V CC
k i
Xi CC k N
= =
=
(4)
( , ) ( , )
, , , , 0
j i E i j E
Xj i k Xi j k
e e
=

, , k i N i k e = (5)
, , , , 0 Xi j k Xj CC k = , j SS k N e e (6)
, , ,
k N
bkXi j k Bi j
e
s
( , ) i j E e (7)
, , 0 Xi CC k Yi s , i k N e (8)
, , , 0 Xi j k Si j s ( , ) , i j E k N e e

(9)
, , , , , {0,1} Xi j k Yi Si j e , , i j k (10)
The main objective is to minimize the cost function given in
the equation (1). The model consists of two types: installation
cost and operation cost. In (1) the cost is the sum of
operational cost at t and onetime cost of installing cellular
towers. End to end latency is restricted in (2). And (3)-(6)
gives the flow conservation constraints. In (7) the total flow
on each link must not exceed the bandwidth. And (8) shows
the link of type (k, CC). (9) Gives cost of link (i, j).
Equation (10) provides decision variables are binary variables.

Fig. 3 Placement graph
VI. SIMULATION RESULTS
To simulate my studies, I use the network simulator 2. This is
a discrete event simulator. The results and discussions are
made as follows.
A. Wired Scenario: In this section the one level of the hybrid
architecture is designed. The description of nodes in the
simulator is given as node 1-Control centre, 2-Substation 1, 3-
Substation 2, 0-Base station. The wired scenario explains the
concept of wired communication between the substations and
control centre. The communication is enhanced by means of
data packets sending between the nodes. And disturbance in
any node is explained by means of packet dropping in the
nodes. These are clearly mentioned in the following
simulation results. Fig. 4 shows the data transfer between
nodes 2, 0 and1. And also shows the acknowledgement
coming back to the node 2 from node 1. Fig. 5 shows the
dropping of packets due to some disturbance in the nodes.
B. Wireless Scenario: In this section, the communication
between the transmission towers and cellular towers are
enhanced by means of creating the wireless nodes. Here the
placement of cellular towers enables the coverage of two or
more transmission towers. The description of nodes in
wireless scenario 0&2-Cellular towers 1, 3, 4, 5, 6, 7, 8-
Transmission towers. Fig. 6 shows the placement and
coverage area of transmission towers and cellular towers.
Each cellular node will cover nearly three transmission nodes.
C. Wired cum Wireless Scenario: In this section, three level
hierarchical architecture is achieved. The wired and wireless
nodes are created and the data transmissions between the
nodes are studied. Fig. 7 shows the wireless node coverage
7
th

area and the communication between wired nodes 1 and2.
There is a flow of data between nodes 1 and 2.

Fig. 4 Data transfer between nodes 2, 0 and 1.

Fig. 5 Dropping of packets.

Fig. 6 Placement and coverage area of cellular and
transmission nodes.

Fig 7 Communication between wired and wireless nodes.
VII. CONCLUSION
The smart grid of the future is generally characterized by
more sensors, more communication, more computation, more
control, but a comprehensive conceptual architecture is seldom
presented. The assumption of a certain generic configuration of
more sensors, more communication, more computers, more
control, from which I try to lay out the total information picture.
From that the objective of how the present applications can be
enhanced and new applications be developed that will make the
operation of the grid more secure and reliable is viewed. Finally,
the layout of a systematic plan of how we can transition from
the present grid to the smart grid is studied. In this work, the
transmission of time sensitive sensor data through the
transmission line network in the presence of delay and
bandwidth constraints are studied. The analysis shows that a
transmission line monitoring framework using WSN is indeed
feasible using accessible technologies. The anticipated
formulation is broad and encompasses variation in several
factors such as asymmetric data creation at towers, wireless
connection reliabilities.
REFERENCES
[1] P. Zhang, F. Li, and N. Bhatt, Next-generation monitoring,
analysis, and control for the future smart control center,
IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 186192, Sep.
2010.
[2] S. Malhara and V. Vittal, Mechanical state estimation for
overhead transmission lines using tilt sensors, IEEE Trans.
Power Syst., vol. 25, no. 3, pp. 12821290, Aug. 2010.
[3] V. C. Gungor, L. Bin, and G. P. Hancke, Opportunities and
challenges of wireless sensor networks in smart grid, IEEE
Trans. Ind. Electron., vol. 57, no. 10, pp. 3557-3564, Oct.
2010.
[4] A. Bose, Smart transmission grid applications and their
supporting infrastructure, IEEE Trans. Smart Grid, vol. 1,
no. 1, pp. 1119, Jun. 2010.
[5] R. A. Leon, V. Vittal, and G. Manimaran, Application of
sensor network for secure electric energy infrastructure,
IEEE Trans. Power Del., vol. 22, no. 2, pp. 10211028,
Apr. 2007.
[6] Ye Yan, Yi Qian, Hamid Sharif, and David Tipper, A
survey on smart grid communication infrastructure:
motivation, requirements and challenges, IEEE
Communications, Surveys & tutorials, vol.15, no. 1, First
Quarter 2013.
[7] P. Ramachandran, V. Vittal, and G. T. Heydt, Mechanical
state estimation for overhead transmission lines with level
spans, IEEE Trans. Power Syst., vol. 23, no. 3, pp. 908
915, Aug. 2008.

7
th

RELIABLE COMMUNICATIONS IN WSN AGAINST GLOBAL
EAVESDROPPER

DINESH KUMAR.V.S GOPINATHAN.B
Pg scholar, Dept of CSE Associate Professor, Dept of CSE
Hosur, Tamil Nadu, India. Hosur, Tamil Nadu, India
dkdineshkumar464@gmail.com gopinathanme@gmail.com

Abstract In device network several
protocols square measure exploitation for privacy
and preservation of knowledge against aggressor.
Such connected data will be manipulate by associate
person to derive sensitive data like the locations of
observe objects and knowledge receivers within the
field. Attacks on these parts will considerably
undermine any network application. The listener, is
realistic and may defeat these existing technique. It
1st formalizes the placement privacy problems in
device networks beneath this sturdy person model
and computes a bound on the communication
overhead required for achieving a given level of
location privacy. It proposes 2 techniques to produce
location privacy to sender-location privacyperiodic
assortment and sender simulationand 2 techniques
to produce location privacy to Receiver-location
privacyReceiver simulation and backbone
flooding. These techniques give trade-offs between
privacy, communication price, and latency. Use of
those propose techniques, it improves location
privacy for each sender and receiver locations.

Index TermsSensor networks, location privacy.

INTRODUCTION
A wireless device network (WSN) usually consists of a large
number of tiny, multifunctional, and resource unnatural
sensors that square measure self-organized as a poster hoc
network to watch the physical world [1]. device networks
square measure typically employed in applications wherever
it's troublesome or impracticable to line up wired networks.
Examples embrace life surround observance, security and
military police work, and target chase.
For applications like military police work, adversaries have
sturdy incentives to snoop on network traffic to get valuable
intelligence. Abuse of such data will cause financial losses or
endanger human lives. to guard such data, researchers in
device network security have targeted significant effort on
finding ways that to produce classic security services like
confidentiality, authentication, integrity, and accessibility.
although these square measure important security needs,

they're inadequate in several applications. The communication
patterns of sensors will, by themselves, reveal an excellent
deal of discourse data, which may disclose the placement data
of important parts during a device network. for instance, in the
Panda-Hunter situation [15], a device network is deployed to
trace vulnerable large pandas during a bamboo forest. every
panda has associate electronic tag that emits an indication that
may be detected by the sensors within the network. A device
that detects this signal, the sender device, then sends the
placement of pandas to an information receiver (destination)
with facilitate of intermediate sensors. associate person (the
hunter) might use the communication between sensors and
therefore the knowledge receivers to find then capture the
monitored pandas. In general, any target-tracking device
network is liable to such attacks. As another example, in
military applications, the enemy will observe the
communications and find all knowledge receivers (e.g., base
stations) within the field. revealing the locations of the
receivers throughout their communication with sensors might
enable the enemy to exactly launch attacks against them and
thereby disable the network.
Location privacy is, thus, vital, particularly in hostile
environments. Failure to guard such data will fully subvert the
meant functions of device network applications. Location
privacy measures, thus, ought to be developed to stop the
person from crucial the physical locations of sender sensors
and receivers. owing to the restricted energy time period of
powered device nodes, these strategies got to be energy
economical.Since communication in device networks is far
costlier than computation [23], It use communication price to
live the energy consumption of our protocols.
Providing location privacy during a device network is
challenging. First, associate person will simply intercept
network traffic owing to the utilization of a broadcasting for
routing packets. He will use data like packet coordinated
universal time and frequency to perform traffic analysis and
infer the locations of monitored objects and knowledge
receivers. Second, sensors typically have restricted process
speed and energy provides. it's terribly costly to use ancient
anonymous communication techniques for concealing the
communication between device nodes and receivers. It ought
7
th

to notice different means that to produce location privacy that
accounts for the resender limitations of device nodes.
Recently, variety of privacy-preserving routing techniques
are developed for device networks. However, most of them
square measure designed to guard against associate person
solely capable of eavesdropping on a restricted portion of the
network at a time. A extremely impelled person will simply
snoop on the whole network and defeat these schemes. for
instance, the person might deploy his own set of device nodes
to watch the communications within the target network [17].
this can be very true during a military or industrial spying
context, wherever the person has sturdy, doubtless crucial ,
incentives to realize the maximum amount data as potential
from observant the traffic within the target network. Given a
world read of the network traffic, the person will simply infer
the locations of monitored objects and receivers. for instance,
an area within the network withhigh activity ought to be near
a receiver, whereas an area wherever the packets originate
ought to be near a monitored object.
Focus on privacy-preserving communication strategies within
the presence of a world listener UN agency includes a
complete read of the network traffic. The contributions during
this paper square measure twofold.

It indicate that the idea of a world listener UN agency
will monitor the whole network traffic is usually realistic for
extremely impelled adversaries. It then formalize the
placement privacy problems beneath such associate
assumption associated apply an analysis supported Steiner
trees to estimate the minimum communication price needed to
attain a given level of privacy.
It give the primary formal study of the way to
quantitatively live location privacy in device networks. It then
apply the results of this study to judge our planned techniques
for location privacy in device networks. These embrace 2
techniques that hide the locations of monitored objects
periodic assortment and sender simulationand 2 techniques
that give location privacy to knowledge receiversreceiver
simulation and backbone flooding. Our analysis and
simulation studies show that these approaches square measure
effective and economical.

EXISTING APPROACHES
Location privacy has been a lively space of analysis in recent
years. In location-based services, a user might want to retrieve
location-based knowledge while not revealing her location.
Techniques like k-anonymity [2] and personal data retrieval
[10] are developed for this purpose. In pervasive computing,
users location privacy will be compromised by observant the
wireless signals from user devices [24], [27]. Random delay
and dummy traffic are prompt to mitigate these issues.
Location privacy in device networks conjointly falls beneath
the final framework of location privacy. The person monitors
the wireless transmissions to infer locations of important
infrastructure. However, there square measure some
challenges distinctive to device networks. First, device nodes
square measure typically battery hopped-up, that limits their
useful time period. Second, a device network is usually
considerably larger than the network in good home or assisted
living applications.
Sender-location privacy: Prior add protective the placement of
monitored objects wanted to extend the safetyperiod, i.e., the
quantity of messages sent by the sender before the item is
found by the aggressor [15]. The flooding technique [20] has
the sender node send every packet through varied ways to a
receiver, creating it troublesome for associate person to trace
the sender. pretend packet generation [15] creates pretend
senders
Whenever a sender notifies the receiver that it's real
knowledge to send. The pretend senders square measure
removed from the $64000 sender and just about at a similar
distance from the receiver because the real sender. Phantom
single-path routing [15] achieves location privacy by creating
each packet walk on a random path before being delivered to
the receiver. Cyclic demurrer [19] creates process ways at
numerous places within the network to fool the person into
following these loops repeatedly and there by increase the
protection amount. However, of these techniques assume an
area listener UN agency is merely capable of eavesdropping
on a tiny low region. a world listener can simply defeat these
schemes by locating the primary node initiating the
communication with the bottom station. Recently, many
techniques are planned to deal with world eavesdroppers.
Receiver-location privacy: In [6], Deng et al. delineated a
method to guard the locations of receivers from an area
listener by hashing the ID field within the packet header. In
[8], it had been shown that associate person will track
receivers by closing time correlation and rate observance
attacks. To mitigate these 2 varieties of attacks, Deng et al.
introduced a multiple-parent routing theme, a controlled
stochastic process theme, a random pretend path theme, and a
hot spots scheme[8]. In [13], redundant hops and faux packets
square measure adscititious to produce privacy once
knowledge square measure sent to the receiver. However,
these techniques all assume that the person may be a native
listener. a world listener will simply defeat these schemes. for
instance, the world listener solely has to establish the region
exhibiting a high variety of transmissions to find the receiver.
It, thus, specialize in privacy protective techniques designed to
defend against a world listener.

NETWORKS AND PERSON MODEL
Sensor networks square measure a comparatively recent
innovation. There square measure variety of various styles of
device nodes that are and still be developed [12]. These vary
from terribly tiny, cheap, and resource-poor sensors like
SmartDust up to PDA-equivalent sensors with ample power
and process capabilities like Stargate. Applications for
7
th

networks of those devices embrace several types of
observance, like environmental and structural observance or
military and security police work.
It think about a homogenous network model. within the
homogenous network model, all sensors have roughly a
similar computing capabilities, power sources, and expected
lifetimes. this can be a typical specification for several
applications nowadays and can seemingly still be widespread
moving forward. it's well studied and provides comparatively
easy analysis in analysis also as straightforward preparation
and maintenance within the field.
Though analysis will be applied to a spread of device
platforms, most sensors flee battery power, particularly within
the varieties of doubtless hostile environments that square
measure learning. Given this, every device includes a
restricted life and therefore the network should be designed to
preserve the sensors power reserves. it's been incontestable
that sensors use way more battery power transmittal and
receiving wireless communications than the other sort of
operation [23]. Thus, focus our analysis on the quantity of
communication overhead incurred by our protocols.
For the varieties of device networks that envision, expect
extremely impelled and well-funded attackers whose objective
is to find out sensitive data like the locations of monitored
objects and receivers.
The objects monitored by the network will be important.
Such objects may be troopers, vehicles, or robots during a
combat zone, security guards during a protected facility, or
vulnerable animals within the wild. If the locations of those
objects were glorious to associate person, the vulnerable
animals may be captured for profit, security guards may be
evaded to alter thieving of valuable property, and military
targets may be captured or killed. Receivers are important
parts of device networks. In most applications, receivers act as
gateways between the multihop network of device nodes and
therefore the wired network or a repository wherever the
perceived data is analyzed. not like the failure of a set of the
sensors, the failure of a receiver will produce permanent
injury to device network applications. Compromise of a
receiver can enable associate person to access and manipulate
all the data gathered by the device network, as a result of in
most applications, knowledge don't seem to be encrypted
when they reach a receiver. In some military applications,
associate person might find receivers and build the device
network nonfunctional by destroying them. Thus, it's typically
important to the mission of the device network to guard the
placement data of monitored objects also as knowledge
receivers.
It think about world eavesdroppers. For a impelled aggressor,
eavesdropping on the whole network may be a quick and
effective thanks to find monitored objects and receivers. There
square measure 2 realistic choices for the aggressor to attain
this. the primary possibility is to deploy his own snooping
device network to listen in on the target network. Note that, at
this value for a BlueRadios SMT Module at $25, the aggressor
wants solely $25,000 to make a network of one,000 nodes [3].
Thus, for even moderately valuable location data, this will be
well worth the price and bother. an alternative choice is to
deploy some powerful nodes to listen in on the network.
However, owing to the short radio ranges of typical device
platforms, the snooping nodes still ought to be deployed
densely enough to sense the radio signals from all device
nodes. Thus, in observe, it's going to not be able to cut back
the quantity of snooping nodes considerably by exploitation
powerful devices. Overall, It think about the primary
possibility as additional sensible for the person.
it's definitely potential that associate person deploys sensors
to directly sense the objects of his interest, rather than
collection and analyzing the traffic within the original
network. However, directly recognizing associate object may
be a terribly difficult drawback in observe owing to the issue
of identifying the physical options of the objects from
background noises. for instance, recognizing a panda is far
tougher than detection a packet and estimating some physical
options (e.g., RSSI) from this packet. In most eventualities,
such sensing drawback is just avoided by putting in atiny low
device (e.g., a device node) on every object; these tiny devices
emit signals from time to time in order that it will sense them
accurately. Thus, locating objects by observance the traffic
within the original network becomes far more engaging to the
person. It think about our defense successful if the person is
forced to launch the direct sensing attack.
though such associate eavesdropping device network would
face some system problems in having the ability to report the
precise temporal arrangement and placement of every target
network event, don't believe that these would keep the
attackers from learning additional approximate knowledge
values. this type of aggressor would be able to question his
own network to work out the locations of determined
communications. He might have acceptable sensors that send
signals that might then be physically placed. He might equip
his sensors with GPS to urge locations or use localization
algorithms to avoid the value of GPS [25], [18]. It don't
assume that the person has got to exactly find every node
within the target network. In most cases, a rough plan
concerning wherever the important events occurred would be
sufficient for the person.
It should, thus, be possible to watch the communication
patterns and locations of events during a device network via
world eavesdropping. associate aggressor with this capability
poses a big threat to location privacy in these networks. It,
therefore, focus our attention to the present sort of aggressor.

Sender-Location Privacy
Periodic assortment
The analysis in Section five shows that the periodic
assortment technique achieves best location privacy.
additionally, the communication overhead within the network
7
th

remains constant and is freelance of each the quantity of
pandas and their patterns of movement. Hence, the main target
of our simulation analysis is on the latency and therefore the
packet drop rate once there square measure multiple pandas
within the field.It set the measure for periodic assortment.are
multiple pandas. It will see that because the variety of pandas
will increase, the latency will increase. this can be as a result
of the nodes near the bottom station receive multiple reports at
a similar time, which needs them to buffer the packets. once
the quantity of pandas grows overlarge, the buffered packets
begin being born owing to the restricted size of the queue, and
therefore the latency of the packets that do hit the bottom
station becomes stable when a definite purpose. once the letter
of the alphabetueue size q decreases, packets traveling long
distances have a high likelihood of obtaining born, creating
the latency of the packets that do hit the bottom station
smaller. this will be seen by a come by the latency for smaller
values of letter of the alphabet within the figure.
It shows the share of the detected events received by the
bottom station. It will see that the share of events received
decreases once there square measure additional pandas within
the field. Increasing letter of the alphabet will definitely
increase the share of the events forwarded to the bottom
station. However, when a definite purpose, increasing letter of
the alphabet won't considerably raise the packet drop rate, as
seen by the tiny distinction from once letter of the alphabet =5
to letter of the alphabet = twenty. On the opposite hand, it tend
to see from Fig. three that increasing letter of the alphabet can
considerably increase the latency of packet delivery.Thus,
fairly tiny values of letter of the alphabet can typically gift the
most effective trade-off purpose between packet drops and
latency. Overall, the ends up in Figs. three and four provides a
guideline for configuring the letter of the alphabetueue size q
to satisfy numerous needs.

Sender Simulation
According to the analysis, the placement privacy achieved by
the sender simulation approach is set by the quantity of virtual
senders simulated within the network. Thus, the main target of
our simulation analysis is on what proportion communication
price we've got to pay to attain a given level of location
privacy. we tend to use these results parenthetically the
potency of the planned technique. throughout the simulation,
we tend to assume that there's only 1 panda within the
network. Multiple pretend pandas square measure created and
simulated within the field. The initial positions of the pretend
pandas square measure indiscriminately chosen. additionally,
assume that the device network is deployed to handle period
applications. In alternative words, whenever a device node
receives a packet, it'll forward it to successive hop as shortly
as potential. Thus, whereas we tend to set the measure for
periodic assortment as, we tend to set it to ten for sender
simulation. In alternative words, in sender simulation, nodes
can forward packets 10 times quicker than within the periodic
assortment technique. It implies that the person has a similar
knowledgeabout the panda behavior because the defender and
therefore cannot distinguish between pretend pandas and real
pandasbased on the determined behavior. It shows the
communication overhead concerned in sender simulation
technique to attain a given level of privacy. It will see that the
communication overhead will increase because the location
privacy demand will increase. This figure conjointly includes
the performance of alternative approaches for any comparison.

Comparison
currently compare the planned source-location privacy
approaches during this paper with 2 alternative privacy-
preserving techniques: phantom single-path routing [15] and
proxy based mostly filtering [29]. It tend to specialize in the
placement privacy achieved and therefore the communication
overhead introduced within the following comparison. The
overhead of the phantom single-path routing theme is
diagrammatic by a single purpose at the bottom-left corner of
the figure, and overheads of the periodic assortment and
therefore the proxy based mostly filtering techniques square
measure diagrammatic by points on the proper a part of the
figure.
In terms of privacy, It've got already shown that none of the
previous strategies (including phantom single-path routing)
will give location privacy beneath the idea of a world listener.
In distinction, each of our strategies give location privacy
against a world listener. The periodic assortment technique
provides the very best level of privacy and is appropriate for
applications that collect knowledge at an occasional rate and
don't need period knowledge delivery, whereas the sender
simulation technique will support period applications with
sensible trade-offs between privacy, communication overhead,
and latency.
It shows the communication prices concerned indifferent
strategies. The simulation results square measure as It might
predict from intuition. The phantom single-path routing
technique introduces comparatively very little communication
overhead, whereas the amountic assortment technique
involves vital hoItver constant communication price for a
given period of your time. The sender simulation technique
provides increasing levels of privacy at the value of additional
communication. It tend to notice that within the figure, the
periodic assortment technique needs less communication
overhead to attain privacy of around b=12 bits in comparison
with the sender simulation technique. the explanation is that
the sender simulation technique is organized to support period
applications with a measure tenth part the length of that
employed in the periodic assortment technique.
It notice that the value of the proxy-basedfiltering (PFS)
technique [29] lies between the prices of the periodic
assortment technique and therefore the (theoretical) Steiner
tree-based technique. However, each of our strategies even
have blessings over PFS. First, throughout simulation of PFS
7
th

technique, it noticed that around seventy p.c of events were
received by the bottom station. However, for the periodic
assortment technique, the detection rate will be as high as
ninety nine p.c. . Second, the sender simulation theme will
give sensible tradeoffs between location privacy and
communication price. It will clearly see that the sender
simulation plan can do a more robust detection rate once the
privacy demand is b=6 or fewer bits.
It also can see the performance of those techniques in
comparison to the approximate Steiner tree formula. For
achieving the most privacy, the periodic assortment technique
consumes additional energy than the approximate Steiner tree
formula. the explanation is that, within the periodic assortment
theme, every device emits a packet each seconds, whereas
within the approximate Steiner tree formula, every device
emits a packet once each seconds, as is that the case with a
true sender .

Receiver-Location Privacy
Receiver Simulation
The analysis within the location privacy achieved and
therefore the quantity of energy consumed by the receiver
simulation theme rely upon the quantity of faux base stations
simulated within the network. The packets generated by the
senders are sent to all or any fake and real base stations.
Hence, the main target of our simulation analysis is on the
latency and therefore the packet drop rate once there square
measure multiple base stations within the field. Fig. seven
shows the latency of packet delivery once thereare multiple
pretend base stations within the field. It will see that because
the variety of faux base stations will increase, there by
providing additional location privacy, the latency will
increase. this can be as a result of having additional base
stations causes additional traffic within the network and
therefore additional packets to be buffered. once the quantity
of faux base stations grows overlarge, the buffered packets
begin being born owing to nodes restricted queue sizes,
whereas the latency of the packets that do hit the bottom
station becomes stable when a definite purpose. once the letter
of the alphabetueue size q decreases, packets traveling long
distances have a high likelihood of obtaining born, creating
the latency of the packets that do hit the $64000 base station
smaller. this will be seen by a come by the latency for smaller
values of letter of the alphabet. It shows the share of detected
events receivedby the $64000 base station. It see that the share
of events received decreases once there square measure
additional pretend base stations within the field. It offer
pointers for configuring the letter of the alphabetueue size q
and therefore the variety of faux base stations to satisfy
numerous needs.
Backbone Flooding
The location privacy achieved by the backbone flooding
approach will increase with the quantity of backbone
members. Packets generated by a sender square measure sent
to all or any backbone members. Hence, the main target of our
simulation analysis is on the delivery latency, the packet drop
rate, and therefore the energy needed for backbone creation.
It shows that increasing the backbone size can cause
additional energy to be consumed. It conjointly see that a rise
within the parameter m, the mincover, can result inmore
backtracking within the backbone creation and thus consume
additional energy.
It shows that the latency of packet delivery will increaseas
the dimensions of the backbone increases. this can be as a
result of a rise within the backbone size can cause a rise inthe
variety of packets within the network, inflicting buffering of
additional packets and a corresponding increase in latency.
It shows the share of the detected events received by the
bottom station. It will see that the share of events received
decreases once there square measure additional backbone
members within the field. It got to build trade-offs between
the latency and therefore the packet drop rate to satisfy
numerous needs.
Comparison
It value the planned receiver-location privacy approaches. It
specialize in the placement privacy achieved and therefore the
communication overhead introduced by every technique. The
simulation results areshown in Fig. 12.
In terms of privacy, it have already shown that none of the
previous strategies will give location privacy beneath the idea
of a world listener. In distinction, each of strategies give
receiver-location privacy against a world listener.
It compare the communication overheads through
simulation. Fig. twelve shows the communication prices
concerned in several strategies. each techniques will give
sensible trade-offs between privacy and communication price.
It note that backbone flooding consumes less energy. the
explanation is that this technique doesn't incur a lot of price to
get traffic toward the pretend base stations. one broadcast of
packets within the backbone effectively creates several
pretend base stations. It note that each the approximate Steiner
tree and backbone flooding techniques square measure
support curves as a result of one packet transmission will be
received by all neighbors of the sender. All of the neighbors
are thought of by the person to be equally seemingly to be a
true base station. Hence, the energy consumption can stay a
similar for privacy within the vary.
In see the result of multiple real base stations on
communication price for the required level oflocation privacy.
every sender sends each packet to each base stations. It
indiscriminately placed the 2 base stations within the network.
The communication price of backbone flooding doubles once
the quantity of base stations doubles. this can be as a result of,
by design, the sender communicates with every backbone
severally. However, the Steiner tree formula solely incurs
atiny low increase in communication price. It will see that
once build the approximate Steiner within the case of multiple
base stations, the communication price remains constant till
7
th

the privacy demand grows on top of seven bits. this can be as
a result of the packets from a sender can continuously bear a
similar ten hops and these ten hops cowl as several sensors as
needed for concerning seven bits of privacy.
Discussion on exploitation the planned Techniques
The planned location privacy techniques during this
paperhave blessings and downsides in comparison with one
another. It concisely summarize our understanding of that
solutions ought to be used for various applications. The
periodic assortment and sender simulation strategies will be
used for providing sender-location privacy. The periodic
assortment technique provides the very best location privacy
and is thus helpful once observance extremely valuable
objects. in addition, the communication costthough high
does not increase with the quantity of monitored objects.
Thus, it's appropriate for applications that collect knowledge
at an occasional rate from the network concerning several
objects. The sender simulation technique provides a trade-off
between privacy and communication prices. it's appropriate
for eventualities wherever 1) the item movement pattern will
be properly shapely and 2) ought to collect period knowledge
from the network concerning the objects.
The receiver simulation and backbone flooding strategies
will give location privacy for the receivers. The backbone
flooding technique is clearly additional appropriate for the
cases wherever a high level of location privacy is required, as
It will see from Fig. 12. However, once the specified level of
location privacy is below a definite threshold (e.g., 6.4 bits as
shown in Fig. 12), the receiver simulation technique becomes
additional engaging, since it's additional sturdy to node failure
within the network. within the backbone flooding plan, It
ought to continuously keep the backbone connected and
construct the backbone from time to time to balance the
communication costsbetween nodes.

CONCLUSIONS
previous work on location privacy in device networks
assumed an area listener. This assumption is impossible given
a well-funded, extremely impelled aggressor. within the
location privacy problems beneath a world listener and
calculable the minimum average communication overhead
required to attain a given level of privacy. It conjointly
conferred techniques to produce location privacy to things and
receivers against a world listener. It used analysis and
simulation to point out however well these techniques perform
in managing a world listener. There square measure variety of
directions that value learning within the future. It assume that
the world listener doesn't compromise device nodes. However,
in observe, the world listener is also able to compromise a set
of the device nodes within the field and perform traffic
analysis with extra information from insiders. It presents
fascinating challenges to our strategies. Second, it takes time
for the observations created by the adversarial network to
succeed in the person for analysis and reaction.

7
th

Abstract In past years, Patient observation is done
manually or by using wireless Body Sensor Network which
is sensibly observed by medical organization agents. Mesh
network is used for reading physiological framework and
clear description of the patient using wireless sensors.
Inventive Agents are proposed for alerting medical
organization and data aggregation. Cloud is also proposed
for supporting healthcare community and remote or
mobile patient monitoring.

I ndex TermsSensor Networks, Patient Observing, Agent
Technology and Cloud Computing.
INTRODUCTION
In the past, Healthcare has been the focus of
many research activities. The project is based on the use of
Information and Computing Technology (ICT) to improve
efficiency in medical, technical and administrative processes.
Patient monitoring is important to care in emergency rooms,
operation room, critical care and intensive care unit And also
invaluable for recovery rooms, respiratory therapy, transport
out-patient care, cath labs, radiology, ambulatory, home
gastroenterology departments and sleep application. Many
problems occur within this and issues of patient observing.
Patient monitoring is a critical function because patient under
medical observation can change in any time. In critical cases,
ICT enables significant reduction of the possibility of human
error. Patient observation is done manually by capture the
physiological conditions of the patient such as pulse rate,
temperature and blood pressure etc. The patients readings are
recorded on the medical chart provided for patients and the
treatment plan is based on captured data.ICT not only make
the automation of the patient observing process possible and
also significantly improve the process. In this paper, we
propose a solution to this problem .From the generic nature of
the solution that we understand it can be applied in many
other situations. This paper gives an overview of Integrating
wireless sensor network with cloud computing, which is as
follows: Section 2 discusses about WSN and Cloud
computing, section 3 discusses about proposed system,
section 4 and 5discusses about simulation scenarios and
simulation results , section 6 discusses concluding remarks.

II.WIRELESS SENSOR NETWORKS:

Wireless Sensor Networks have originated
as a vital new area in wireless technology. Initially Sensor
Networks were developed only for military applications such
as battlefield monitoring and have been successfully retreated
for patient monitoring backbone network which creates.
The sensor network model is a database model. The term
computer network model defines the category in which a
computer network can be grouped into. This network models
are possibly still the most important of the special structures in
linear programming. The network users hardware or software
in the share way over the network and this sensor model
clearly defines the functions of communication software in a
generalized and structured manner which helps to carry out
the network product development activities. The approach
presented here is simply derived from specializing the rules of
the simple method to take advantage of the structure of
network models. All WSNs are controlled by software which
implements the different routing protocols used by the
network.

III. CLOUD COMPUTING:

Cloud computing is used to describe a
variety of different types of computing and a large number of
computers is connected through Internet. It is based on Pay-
Per-Use services .In cloud there are three basic services
available they are Software as a Service , Platform as a
Service and Infrastructure as a Service. But health care service
uses only a good Internet connection. This service enables
small healthcare hospital to multi-specialty hospital to pay per
use service which is cloud service as like paying for Internet
connection service. Based on a cloud based hospital
management, it uses application program interface which
connects emergency ward workers with pre stored data and
connected to ambulances. Then Doctors are allowed to see
important data which is collected in ambulances through the
hospitals Emergency ward and fed it into patients Electronic
Wireless Body Sensor Network Integrating with Cloud computing
for Improving the Patient Observation

S. Janani Devi
*1
,Dr.G.M.Tamilselvan
*2
, M. Suresh
*3

*1&*3
PG Scholar, Bannari Amman Institute of Technology
*2
Associate Professor,BannariAmmanInstituteofTechnology
mail2jananis@gmail.com,sureshm.co12@bitsathy.ac.in
7
th

Health Record. In before projects the Emergency ward doctors
had to fax the patient data manually.

IV THE PROPOSED SYSTEM:

Integration of Wireless Body Sensor
Network integrating with Cloud Computing are proposed for
more efficient patient observing. Currently healthcare centers
use Wireless Body Sensor Networks (WBSNs) to observe the
patients and normally WBSNs formed in an ad-hoc
environment, which bring frequent network failures. Wireless
Body Sensor network is proposed in this method.WBSN
provides different functionalities to improve the monitoring of
environment. It uses wireless sensors for reading
physiological parameters and patient identification. And also
we are storing the result obtained in QualNet software in the
cloud storage device. And also the main contribution of this
paper based on integrating the wireless sensor network which
is integrated with Cloud computing which would see the
patients chart and tshe agent can be programmed to perform
some serving job on the sensed data, which lead to reduction
in the network traffic so reducing network response time. The
system will support patient observing models which are
described by Bayesian classifiers and accept the training of
agents to make determination by intelligent over the variations
in required readings of observed parameters and the original
readings. Therefore a Cloud technology is proposed which is
used to represent a Community Cloud. The organization of
cloud is under the control of multiple organizations which
deal some common interest as like health care facilities.

Fig 1 Proposed Architecture
In figure1, it shows the proposed system architecture and its
various components. The system architecture has four agents
they are Aggregator Agent, Patient Agent, Doctor Agent and
Nurse Agent. A group of patients has been connected to the
cluster head which act as Patient Agent. Different cluster head
is used to send the information to the Access point which is
used to access the patient information. The Patient agent is
situated at the cluster head of the network and the information
has been transferred to the base station where Aggregator
Agent presents. The Aggregator Agent is used to receive the
information and checks for denotations of anomalous readings
which is sent by the Patient Agent and start alerting to Doctor
Agent and Nurse Agent. Then the Aggregator Agent transmits
the patients parameters to the Cloud computing for
destination processing and storing in the database. The
Doctor Agent sand Nurse Agent are situated on the
mobile handheld device and send the information for the on
duty Doctor and nurse and also to the patients assigned
doctor who may not be on duty. The Doctor Agent and Nurse
Agent offer alerts to the medical agents and allow the
querying of patients information which includes current and
past sensor readings .The proposed system will be extended to
support mobile or remote patients and using Patient Agents
which is situated on mobile handheld devices capable of
receiving and transmitting readings for the sensors observing
the patients physiological information. By using cloud
computing, the information from the Patient information has
been stored in the cloud storage device. The Proposed system
can be considered as a Medical Internet of Things which is
possible to observe, track and uniquely identify all these
information which is connected to the system via the Internet.

V SIMULATION SCENARIOS:

Fig 2 Data transfer between Access points

In this figure 2, the group of nodes has been connected to the
cluster head. The information from the group of patients has
been transferred to the cluster head which act as a patient
Agents. And then from the three different cluster heads, the
information has been transferred to the base station where the
aggregator agents present to receive the information. From the
7
th

Aggregator Agent, the information has been transferred to the
doctor agent and nurse agent through another Access Point.
The information has been transferred from one access point to
the other access point and then it transfer to the particular
agent which belongs to the information of the patient.

Fig 3 Data transfer between access point and cluster head

In this figure 3, Data transfer from one cluster head to other. It
means the group of patients information has been transferred
from one agent to other agent. Then the agent sends the
information to the access point which is used to access the
information of the patient and send to the base station. From
the base station, the data has been transferred to another
access point which the information is then passed to the nurse
agent and doctor agent.

VI. SIMULATION RESULTS

Fig 4: Average throughput

Throughput means sum of data rates that are delivered to all
terminals in a network. Average rate of successful message
delivery. It passes through certain network node. Measured in
bits/second. Graph shows the information which is received at
the receiver. Node 1, 3 and 5 are cluster heads which receives
information from the patients. And the information received is
shown in above graph.

Fig 5: Average Delay
In this figure 6, the average delay shows the information
transferring delay time in the simulation process. Delay varies
on each and every node due to the variation of packet size.
And it only occurs between 0 to 3.5. Nodes 1, 3 and 5 are
cluster heads in which delay occurs. Delays occur in the
cluster heads which is also called as a patient Agent. Metric
value of different nodes is used to determine the best possible
route which has low metric value.

Fig 6: Received information
Received information such as delay, throughput and jitter
value from the three different cluster heads is received. The
node 1,3 and 5 are cluster heads which sends the information
to the access point of node 14.Metric value is to determine the
best possible route which has low metric value.
CLOUD STORAGE SIMULATION

Fig 7: Simulation of Net beans software

In this Net beans software, the java coding is applied for the
cloud storage. In the front end process, Java software is used
and in the back end, SQL server is used to store the data. The
7
th

coding is applied in this net bean and run the software in the
name applied.

Fig 8: Uploading a file

Then choose the file to be store in the cloud storage device
and upload it manually. Then the uploaded data will be stored
in the cloud to access the information and enables small
healthcare clinics to multi-specialty hospital to pay per (cloud)
service, similar to paying for Internet service.

SQL BROWSER:

Fig 9: SQL Browser
The data or result we are getting through the QualNet
software is stored in the cloud or data base server. To store in
the cloud or database, SQL server is used and to access into
cloud storage. The SQL query browser is used to access into
the cloud using particular unique id such as username and
password.

VII.RESULT OF CLOUD STORAGE DATA:

Fig 10: Cloud storage data
In this we are selecting our data which we are going to store in
the database. Select the fileup folder in connection root. And
select the data name which is to be stored in the system. Then
run the simulation. Then the result has been stored in the
cloud storage device after the simulation process. The result
will be stored in the device in the name which we given in the
software

VIII.CONCLUSION:
Integration of wireless body sensor network integrated with
cloud computing is concluded that sensing network can
uniquely identify the patient. And data collected from the
patient, in addition to determining the patient location within
the network and monitoring the patients condition. Cloud is
proposed to store the patient information in it and is based on
Pay per use services. This enables small healthcare clinics to
multi-specialty hospital to pay per (cloud) service, similar to
paying for Internet service

REFERENCES
[1] Norman A. Benjamin1 and Suresh Sankaranarayanan
University of West Indies, Kingston, Jamaica
Performance of Wireless Body Sensor based Mesh
Network for Health Application International Journal of
Computer Information Systems and Industrial
Management Applications.ISSN: 2150-7988 Vol.2 ,
pp.020-028(2010).
[2] Miller K and Suresh S, Role of Intelligent Agents and
Wireless Body Sensor Mesh Networks in Patient
Monitoring, Presented in the 5th IEEE International
conference in Networked Computing, Seoul, Korea, 25-
27, August (2009).
[3] Ahsan K., Shah. H , and Kingston. P RFID Applications:
International Journal of Computer Science Issues , Vol. 7,
No. 1, pp 1-7(2012, January).
[4] Benjamin N and Suresh S, Performance of Hierarchical
Agent based Wireless Sensor based Mesh Network for
Health Care-Proceedings of the 8th IEEE International
conference on Computer Information Systems and
Industrial Management, Coimbatore, Chennai, 9-
11,.Pp.1653-1656. ISBN: 978-1-4244-5612-3 December
(2009).
7
th

[5] Benjamin, N. A., Performance of Wireless Body Sensor
based Mesh Network for Health Application,
International Journal of Computer Information Systems
and Industrial Management Applications, pp 21-
28(2009).
[6] Barnes D et al, Performance analysis of Client/server
versus Agent based communication in Wireless Sensor
Networks for health applications, Proceedings of 2009
IEEE Electro Information Technology (EIT 09), Windsor,
Canada, June 7-9, pp.271-276. ISBN: 978-1-4244-3355-
1(2009).
[7] Bui. N., Castellani, P., Rossi, M.,., Zorzi, M. & Shelby M
Architecture and protocols for the internet of things: A
case study. Proceedings of IEEE International
Conference on Pervasive Computing and
Communications Workshops, pp. 678-683(2010).
[8] Sankaranarayanan & Edwards T. Applications of
Intelligent Agents in Hospital Search and Appointment
System. International Journal of E-Services and Mobile
Applications, Vol. 3, No. 4, pp 57-81.(2011).
[9] Shah M.A, Swaminathan R, and Baker M, Privacy
preserving audit and extraction of digital contents,
Cryptology ePrint ,Report 2008/9186, (2008).
[10] Fifah, B S. NFC Enabled Patient appointment System,
Bachelor of Internet Computing Project, Department of
Computing and Information Systems,(2013)
[11] Korkmaz. I., Atay, C., and Kyparisis.G (2010). A
Mobile Patient Monitoring System Using RFID.
Latest Trends on Computers, Vol. 3, pp 726-732.
[12] G. Ateniese, S. Kamara, and J. Katz, Proofs of
storage from homomorphic identification protocols,
in Asiacrypt,pp. 319333(2009).
[13] Kailing .R, Rantzau. S, K. Beier and Grandison, T., ,
(2006). Discovery Services Enabling RFID
Traceability in EPC global Networks. Proceedings of
the 13th International Conference on Management of
Data 2006, Delhi, India. December 2006 .
[14] Mitrokotsa. A, and Douligeris, C Integrate RFID
and Sensor Networks: Architectures and
Applications, pp 511-536. (2009).
[15] Lefebvr.E, Castro. L., and Lefebvre, L. A..
Assessing the prevailing implementation issues of
RFID in healthcare: A five-phase implementation
model. International Journal of Computers and
Communications, Vol. 5, No. 2, pp 101-117(2011).
[16] Morak, J., Schreier, G. MHealth based on NFC
Technology-Preliminary results from Medium Scale
Proof of Concept Projects, Proceedings of Ehealth ,
Austria, (2012).

7
th

INCREASE THE EFFICIENCY IN WIRELESS SENSOR
NETWORK USING MOBI-CLUSTER ALGORITHM

Sridhar.S. , Vidya.A.R.
Department of Computer science and Engineering Regional centre, Anna University-Coimbatore
becsesridhar7@gmail.com , vidyaradhakrishan@gmail.com

Abstract: A wireless sensor network consists of
spatially distributed autonomous sensors to
monitor physical or environmental conditions and
to cooperatively pass their data through the
network to a main location. Utilizing Mobi-
clustering algorithms to form a hierarchical
network topology is a common method of
implementing network management and data
aggregation in wireless sensor networks. Mobi-
clustering algorithm is used minimizing the
overall network overhead and energy. Power
consumption and maximizing the life time are the
important aspects of wireless sensor networks.
The proposed protocol aims at minimizing the
overall network overhead and energy Distribution
associated with the multi-hop data retrieval
process while also ensuring balanced energy
consumption among Sensor Nodes and extended
network lifetime. Arranging cluster sizes and
transmission ranges (ACT), reduces the size of
clusters near the base station and provides energy
consumption. But it does not concentrate on
coverage problem and maximizing the life time. In
this project is developed similar to, the Integer
Linear Programming (ILP) problem that
formulates the coverage problem and maximizes
life network time of the sensor nodes. It
implement the optimization problem, with the
objective function and several constraints of this
problem can be solved to optimality by using
CPLEX solver.

I ndex Terms: Cluster, Energy consumption,
Energy-Balancing, Routing protocol, Wireless
sensor networks.

I. INTRODUCTION

A. Overview Of Wireless Sensor Networks

Wireless Sensor Networks have emerged as
research areas with an overwhelming effect on
practical application development. They permit fine
grain observation of the ambient environment at an
economical cost much lower than currently possible.
In hostile environments where human participation
may be too dangerous in sensor network which may
provide a robust service. Sensor networks are
designed to transmit data from an array of sensor
nodes to a data repository on a server. WSN has

potential to design many new applications for
handling emergency, military and disaster relief
operations that requires real time information for
efficient coordination and planning. Sensors are
devices that produce a measurable response to a
change in a physical condition like temperature,
humidity, pressure etc.

WSNs may consist of many different types of sensor
such as seismic, magnetic, thermal, visual, infrared,
and acoustic and radar capable to monitor a wide
variety of ambient conditions. Through each
individual sensor may have severe resource constraint
in terms of energy, memory, communication and
computation capabilities; large number of them may
collectively monitor the physical world and process
the information on the fly environment. In a WSN,
sensor nodes monitor the environment, detect events
of interest, produce data and collaborate in
forwarding the data towards a sink, which could be a
gateway, base station, storage node, or querying user.
A sensor network is often deployed in an unattended
and hostile environment to perform the monitoring
and data collection tasks. When it is deployed in such
an environment, it lacks physical protection and is
subject to node compromise. After compromising one
or multiple sensor nodes, an adversary may launch
various attacks to disrupt the in-network
communication.

B. Wireless Sensor Networks Characteristics

A WSN is different from other popular
wireless networks like cellular network, WLAN and
Bluetooth in many ways. Compared to other wireless
networks, a WSN has much more nodes in a network,
distance between the neighboring nodes is much
shorter and application data rate is much lower also.
7
th

Due to these characteristics, power consumption in a
sensor network will be minimized. To keep the cost
of the entire sensor network down, cost of each
sensor needs to be reduced. It is also important to use
tiny sensor nodes. A smaller size makes it easier for a
sensor to be embedded in the environment it is in.
WSNs may also have a lot of redundant data since
multiple sensors can sense similar information. The
sensed data therefore need to be aggregated to
decrease the number of transmission in the network,
reducing bandwidth usage and eliminating
unnecessary energy consumption in both transmission
and reception.

The main characteristics of a WSN include,

Power consumption using batteries
or energy harvesting
Ability to cope with node failure
Mobility of nodes
Heterogeneity of nodes
Scalability to large scale deployment
Ease of use

C. Benefits Of Wireless Sensor Network

The two primary motivations for choosing a
wireless network over a wired approach are the
flexibility and the cost-savings associated with
eliminating cables and wires.

Flexibility
Low cost
Lifetime maximization

D. Overview Of Clustering

The large-scale deployment of wireless
sensor networks (WSNs) and the need for data
aggregation necessitate efficient organization of the
network topology for the purpose of balancing the
load and prolonging the network lifetime. Clustering
has proven to be an effective approach for organizing
the network into a connected hierarchy. In this article,
we highlight the challenges of clustering in a WSN,
discuss the design rationale of the different clustering
approaches, and classify the proposed approaches
based on their objectives and design principles. We
further discuss several key issues that affect the
practical deployment of clustering techniques in
sensor network applications. In order to support data
aggregation through efficient network organization,
nodes can be partitioned into a number
of small groups called clusters. Each cluster has a
coordinator, referred to as a cluster head, and a
number of member nodes.

Clustering results in a two-tier hierarchy in
which cluster heads (CHs) form the higher tier while
member nodes form the lower tier. The member
nodes report their data to the respective CHs.
Research on clustering in WSNs has focused on
developing centralized and distributed algorithms to
compute connected dominating sets. The CHs
aggregate the data and send them to the central base
through other CHs. Because CHs often transmit data
over longer distances, they lose more energy
compared to member nodes. The network may be
clustered periodically in order to select energy-
abundant nodes to serve as CHs, thus distributing the
load uniformly on all the nodes. Besides achieving
energy efficiency, clustering reduces channel
contention and packet collisions, resulting in better
network throughput under high load.

E. Cluster Network Characteristics

A basic cluster has the following
characteristics:

Multiple computing nodes
Low cost

A fully functioning computer with its
own memory, CPU, possibly storage
Own instance of operating system
Computing nodes are connected by
interconnects
Typically low cost, high bandwidth
and low latency
Permanent, high performance data
storage

A resource manager to distribute and
schedule jobs
The middleware that allows the
computers act as a distributed or
parallel system
Parallel applications designed to run
on it

F. Benefits Of Clusters

Benefits and reasons for popularity of
clusters can be listed as follows:
No expensive and long development
7
th

projects. Building clusters is easy,
compared to building a dedicated
supercomputer.
Price performance benefit: Highly
available COTS products are used.
Flexibility of configuration: Number
of nodes, nodes performance, and
inter-connection topology can be
upgraded. System can be modified
without loss of prior work. Two types
of scaling can be defined.
Scale up: Increasing the throughput
of each computing node.
Scale out: Increase the number of
computing nodes. Requires efficient
I/O between nodes and cost effective
management of large number of
nodes.

II. RELATED WORK

Wireless distributed micro sensor systems
will enable the reliable monitoring of a variety of
environments for both civil and military applications.
In this paper, we look at communication protocols,
which can have significant impact on the overall
energy dissipation of these networks [2]. Based on
our findings that the conventional protocols of direct
transmission, minimum-transmission-energy, multi-
hop routing, and static clustering may not be optimal
for sensor networks, we propose LEACH (Low-
Energy Adaptive Clustering Hierarchy), a clustering-
based protocol that utilizes randomized rotation of
local cluster base stations (cluster-heads) to evenly
distribute the energy load among the sensors in the
network. LEACH uses localized coordination to
enable scalability and robustness for dynamic
networks, and incorporates data fusion into the
routing protocol to reduce the amount of information
that must be transmitted to the base station.
Simulations show that LEACH can achieve as much
as a factor of 8 reduction in energy dissipation
compared with conventional routing protocols. In
addition, LEACH is able to distribute energy
dissipation evenly throughout the sensors, doubling
the useful system lifetime for the networks we
simulated.

Wireless sensor networks (WSNs) are
composed of a large number of inexpensive power-
constrained wireless sensor nodes, which detect and
monitor physical parameters around them through
self-organization. Utilizing clustering algorithms to
form a hierarchical network topology is a common
method of implementing network management and
data aggregation in WSNs [3]. Assuming that the
residual energy of nodes follows the random
distribution, we propose a load-balanced clustering
algorithm for WSNs on the basis of their distance
and density distribution, making it essentially
different from the previous clustering algorithms.
Simulated tests indicate that the new algorithm can
build more balanceable clustering structure and
enhance the network life cycle. A new method is
proposed in this paper to improve Low Energy
Adaptive Clustering Hierarchy(LEACH) by electing
cluster heads according to the residual energy of the
nodes dynamically[4].

A sliding window is set up to adjust the
electing probability and keep stable the expected
number of the cluster heads using two parameters in
this method, one is the initial energy information of
the nodes and the other is the average energy
information of those that have not already been
cluster heads in the network. Meanwhile, the number
of cluster heads which is fixed in the entire network
lifetime in LEACH is modified to be a variable
according to the number of the living nodes.
Simulations show that the improvement for First
Node Dies (FND) and Half of the Nodes Alive
(HNA) is 41% and 36%, respectively over LEACH,
17% and 26% for Low Energy Adaptive Clustering
Hierarchy with Deterministic Cluster-Head Selection
(LEACH-DCHS), 22% and 21% for Advanced Low
Energy Adaptive Clustering Hierarchy (ALEACH).

Networking together hundreds or thousands
of cheap micro-sensor nodes allows users to
accurately monitor a remote environment by
intelligently combining the data from the individual
nodes. These networks require robust wireless
communication protocols that are energy efficient
and provide low latency [5]. In this paper, they
developed and analyze low-energy adaptive
clustering hierarchy(LEACH), a protocol architecture
for micro-sensor networks that combines the ideas of
energy-efficient cluster-based routing and media
access together with application-specific data
aggregation to achieve good performance in terms of
system lifetime, latency, and application-perceived
quality. LEACH includes a new, distributed cluster
formation technique that enables self-organization of
large numbers of nodes, algorithms for adapting
clusters and rotating cluster head positions to evenly
distribute the energy load among all the nodes, and
techniques to enable distributed signal processing to
save communication resources. Our results show that
LEACH can improve system lifetime by an order of
magnitude compared with general-purpose multi-hop
approaches.
7
th

III. PROPOSED WORK

In a WSN, the sensor network is often
deployed in an unattended and hostile environment to
perform data collection tasks. It lacks of physical
protection and is subject to node compromise when it
is deployed in such an environment. An important
objective of any clustering technique is network
connectivity. For intra-cluster communication, a
cluster member communicates with its CH either
directly. Connectivity in this case is a result of the
success of cluster formation. Network density has to
be sufficiently high in order to ensure that enough
gateways are present at the intersection areas between
clusters. After compromising one or multiple sensor
nodes, an challenger may launch various attacks to
disrupt the in-network communication. compromised
nodes drop or modify the packets that they are
supposed to forward.

Utilizing Mobi-clustering algorithms to form
a hierarchical network topology is a common method
of implementing network management and data
aggregation in wireless sensor networks. Power
consumption and maximizing the life time are the
important aspects of wireless sensor networks. The
proposed protocol aims at minimizing the overall
network overhead and energy Distribution associated
with the multi-hop data retrieval process while also
ensuring balanced energy consumption among Sensor
Nodes and extended network lifetime. Arranging
cluster sizes and transmission ranges (ACT), reduces
the size of clusters near the base station and provides
energy consumption. But it does not concentrate on
coverage problem and maximizing the life time. ACT
consists of the cluster formation phase, data
forwarding phase and cluster maintenance phase. The
following conditions are assumed

The positions of BS and sensor nodes are
fixed.

The power of all sensor nodes is the same in
the beginning.

Each sensor node transmits one unit of data
to the CH in each round.
A CH aggregates data and then forwards
them to BS.

Sensor nodes are uniformly distributed in the
sensor field with density dn.
Sensor nodes can adjust power levels. The
maximum power level can be used in
transmitting data to BS directly.
The covered area is a WL rectangle, where
W is the width and L is the length. Note that
the area considered in LEACH and BCDCP
are also rectangular.
The proposed work consist of the list of modules
which describes the network life time and coverage
area in wsn.

Node Creation.
Cluster head selection.
Cluster building phase.
Cluster maintenance phase.

A. Node Creation

In this module, wireless sensor nodes are
created and information collected from the sensor
nodes. Various information from the sensor nodes
collected though multi hop communication. Calculate
the distance from base station to each sensor node
and individual node to node distance also calculate
using the cluster radius .According to the radio
energy dissipation model [4], the energy consumed
by sensor nodes for transmitting k bits of data at a
distance d is:
(,) = () + (,)
,

That is ,
(,) =2 (1)

is the energy used in a sensor node for a
transmitting one bit of message. is the energy
consumed by the transceiver during its transmission
of one bit data through amplifier.

Eq. (1) is simplified into eq. (2):
(,) =(+)2 (2)

Eq. (2 to calculate the radius of the each cluster.

Fig.1. Level structure
7
th

B. Cluster Head Selection

Cluster head is selected by comparing the
remaining power of the nodes. This approach helps to
prolong the lifetime of the nodes. If the power of
sensor node is higher than other sensor nodes in a
group that will be selected as a cluster head. DSBCA
follows a distributed approach to establish
hierarchical structure in self-organizing mode without
central control. DSBCA selects the random nodes to
trigger clustering process first. Then the trigger node
calculates its connected density and distance from the
base station to determine cluster radius and became
the temporary cluster head selection.

C. Cluster Building Phase

DSBCA sets the threshold of cluster size.
The number of cluster nodes cannot exceed the
threshold to avoid forming large clusters, which will
cause extra overhead and thus reduce network
lifetime. When the cluster head node receives Join
message sent by the ordinary node, it will compare
the size of cluster with threshold to accept new
member and update the count of cluster nodes if the
size is smaller than threshold, or reject the request. If
the rejected node has cluster head already, the
clustering process cases. Otherwise, it finds another
suitable cluster to join. Each member node of cluster
maintains a cluster information table, which saves the
HID, HD, SID and other information.

If a node receives transmitting packet in
work, it will update its cluster information table in
that order. For example, the node checks HD in a
newly received packet, if HD is smaller, then it
updates the value of HD in table, with SID updated.
That is to say, it has found a shorter path to cluster
head and sets the new SID as its forwarding node.
There is only a single HID entry in the ordinary node
because it belongs to one cluster head, but the
overlapping cluster node has multiple HID
information entries for different clusters. DSBCA
algorithm avoids the fixed cluster head
scheme(cluster head manages cluster and forwards
data, so it consumes energy faster than other nodes),
with periodic substitution to balance the node energy
utilization.

Data forwarding :

The data forwarding phase consists of intra-
cluster data forwarding and inter-cluster data
forwarding.,

Intra-cluster data forwarding:
After cluster setup, CHs collect
data transmitted from cluster members and
perform data aggregation. If a CH adopts
single-hop for intra-cluster data forwarding,
the sensor nodes farthest from the CH
consume much more energy in larger
clusters. To avoid this, we employ the
concept of the MST to reduce the distance
between the sensor nodes and CHs for data
transmission.

In a network with a high density of
sensor nodes, the transmitted information
may go far before reaching the targeted CHs
if the MST is applied. Therefore, we assign
a hop count H. At the time the data
transmission begins, the data forwarding
from one sensor node to another causes the
value of H to be decreased by one. When
the value H is equal to zero but the data fails
to reach the targeted CH, the sensor node
that holds the data at the moment passes the
data to the CH directly to avoid time-
consuming routing.

Inter-cluster data forwarding:
The inter-cluster data forwarding
refers to transmissions of CHs collected
data from cluster members in the ith level to
the next CH in the (i-1)th level closest to the
BS until the transmitted data reach the BS.

D. Cluster Maintenance Phase

The power of a CH may be exhausted after
going through cluster setup and several data
transmissions. At this time cluster maintenance, such
as picking out a new CH and merging clusters,
should be performed to continue data transmissions
from sensor nodes to the BS. Cluster maintenance is
also required when a CH is away from its original
cluster due to the mobility of sensor nodes. In the
traditional cluster-based routing protocol, the cluster
maintenance phase is very important because the
loads imposed on the CHs are much larger than those
imposed on the sensor nodes; as a result, the power of
CHs may be exhausted quickly. In case the power of
the CH approaches depletion, a new CH is elected. In
ACT, the cluster maintenance phase consists of CH
rotations within a cluster and cross-level data
transmission to the BS.
7
th

CH rotations in a cluster:
Define the threshold of CH power
as T (15% of initial energy). When the
remaining power of a CH is under T, a new
CH is selected from among the sensor
nodes, usually the one closest to the CH
within its cluster group, while a change msg
is broadcast to inform cluster members of
the change of CH.

Cross-level data transmission to BS:
In ACT, clusters in the 1st level are
the smallest in size, and thus, fewer cluster
members are found in the 1st level. If the
sensor nodes within a cluster take turns
serving as CHs, the process may finish
quickly as there are not enough sensor
nodes. When the BS is aware that each
sensor node in the 1st level can no longer
serve as a CH, it broadcasts a message to
allow the CHs in the 2nd level to transmit
data to BS directly.
Let the network topology be
divided into K levels with the CHs in the 1st
level transmitting data to BS. When the
power of each sensor node in the 1
st
level is
exhausted after a while, the CHs in the 2nd
level assume the process of data
transmission (the same for 3rd level, 4th
level,Kth level). In this way, the network
lifetime can be extended.

Cluster head selection, cluster setup and
intra-cluster communications are the fundamental
methods of forming a cluster. Moreover, while ACT
does consider cross-level data transmission to BS
and communication load balance for each cluster
LEACH, BCDCP and MRLEACH do not.

IV. SIMULATION RESULTS AND

PERFORMANCE ANALYSIS

We conduct simulations to study the performance
of the proposed ACT and the other three schemes,
LEACH, BCDCP and MR-LEACH.

A. Simulation setup

We use a combination of the NS-2 [6] and the
Mannasim [7] in simulations.We apply the first order
radio model [5] to evaluate the energy dissipation in
sensor nodes. We average the results based on 300
runs for each scenario. In addition, we are interested
in the following performance metrics: (1) average
energy dissipation; (2) standard deviation of energy
consumption of CHs (utilizing standard deviation to
observe the scattering of values); (3) the number of
sensors alive; and (4) network lifetime (which is
defined as the number of rounds in which the first
sensor node uses up its energy).

B. Simulation results

The average energy dissipation in sensor
nodes in four types of routing protocols, namely
LEACH, BCDCP, MRLEACH and ACT, within a
range of 80 _ 120 m2. The energy dissipation
found in LEACH is greater than that in BCDCP,
MRLEACH and ACT as a whole. This is because
LEACH adopts single-hop communications with
the CH sending its data directly to the BS;
BCDCP, MR-LEACH and ACT utilize multi-hop
communications that require less energy
consumption from each sensor node.

Fig.2. Network lifetime of different node
densities

The network lifetime for different node
densities of 40, 60, 80, 100 and 120 nodes. We
define network lifetime as the number of rounds
before the first sensor node uses up its energy in
the network.

Table.1. The simulation parameters

PARAMETER VALUE
Network size 80 120
2

Base station location (0,40)m
7
th

Number of sensor nodes 96 nodes
Density 1 node/ 100
2

Initial energy of each 0.5J
sensor node
Data packet size 500 bytes
50 nJ/bit

100 pJ/bit/
2

K 3 for ACT ,4 for MR-
LEACH

V. CONCLUSION

Mobi-clustering algorithm can form more
stable and reasonable clusters, and also improve the
network life cycle. Wireless sensor network requires
sensor node to work for a long period of time without
human intervention. Using Integer Linear
Programming(ILP) problem to avoids the coverage
problem and increases the network lifetime, that
cannot be achieved in existing system. The clusters
formed by DSBCA is based on the distance from base
station, distribution of nodes and residual energy
accord with actual network. Hence, it achieves a
better performance even when the number of nodes
changes As a result, the energy consumption
decreases effectively. The cluster structure changes in
each round in LEACH, but DSBCA maintains
relatively stable clustering structure in which
switching of cluster head often occurs in the same
cluster.

REFERENCES

[1]. Wei Kuang Lai, Chung Shuo Fan, Lin Yan Lin,
sep 2011, Arranging cluster sizes and
transmission ranges for wireless sensor
networks, Elsevier, Taiwan, 183(2012) 117-131.

[2]. W.R. Heinzelman, A. Chandrakasan, H.
Balakrishnan, January 2000, Energy-efficient
communication protocol for wireless microsensor
networks, Proceedings of Hawaii International
Conference on System Sciences (HICSS), Hawii,
pp. 110.

[3]. Ying Liao, Huan Qi, and Weiqun Li, MAY
2013,Load-Balanced Clustering Algorithm With
Distributed Self
Organization for Wireless Sensor Networks,
IEEE
Sensors Journal, VOL. 13, NO. 5.

[4]. Aimin Wang, Dailiang Yang, Dayang Sun ,Dec
2011,Clustering Algorithm Based On Energy
Information And Cluster Heads Expectation
For Wireless Sensor
Networks, Computers and Electrical
Engineering 38
(2012) 662671.

[5]. W.R. Heinzelman, A. Chandrakasan, H.
Balakrishnan, Oct 2002, An application-
specific protocol architecture for
wireless microsensor networks, IEEE
Transactions on
Wireless Communications 1 (2002) 660670.

[6]. Network simulator, NS-2.
<http://www.isi.edu/nsnam/ns/>.

[7]. Mannasim.
<http://www.mannasim.dcc.ufmg.br/index.htm>.

7
th

DESIGN OF A HYBRID WIRELESS NETWORK USING
ENHANCED QUALITY OF SERVICE ORIENTED
DISTRIBUTED ROUTING PROTOCOL

T.Murugeswari, M. Aswathy
Department of Applied electronics, Hindustan college of engineering and technology, Coimbatore

Abstract
As wireless communications gains popularity,
significant research has been devoted to supporting
real time transmission with stringent quality of
service requirements for wireless applications. At the
same time, a wireless hybrid network that integrates a
mobile wireless adhoc network and a wireless
infrastructure network has been proven to be a better
alternative for the next generation wireless network.
Quality of service Oriented Distributed routing
protocol enhance the Quality of Service capability of
hybrid networks. In this paper improve the quality of
protocol and develop a protocol Enhanced Quality of
service Oriented Distributed routing protocol. The
Enhanced Quality Of service Oriented Distributed
routing protocol is mobility resilient than Quality of
Enhanced Quality of service Oriented Distributed
routing protocol improves the throughput and
decreased the overhead. Hence this model shows that
routing protocol can provide high Quality of Service
performance in terms of overhead, transmission delay
and mobility resilient. This also increase the energy
efficiency of Quality of service Oriented Distributed
routing protocol and avoid the energy harvesting
problem by using Enhanced Quality of service
Oriented Distributed routing protocol. Hence it leads
to develop a high efficiency hybrid network using
routing protocol.

Key Words: Infrastructure Wireless Network,
Adhoc Mode Network, Hybrid Wireless Network,
Quality of Service Distributed oriented Routing
Protocol.

I. INTRODUCTION

Nowadays, people wish to watch videos, play
games, watch TV and make long-distance
conferencing via wireless mobile devices. The
emergence and the envisioned future of real-time and
multimedia applications have stimulated the need of
high Quality of Service (QoS) support in wireless and
mobile networking environments [1]. The QoS
support reduces end-to-end transmission delay and
enhances throughput to guarantee the seamless
communication between mobile devices and wireless
infrastructures. In concert hybrid wireless networks
(i.e., multi-hop cellular networks) have been proven
to be a better network structure for the next
generation wireless networks [25], and can help to
deal with the inflexible end-to-end QoS requirements
of different applications. Hybrid networks
synergistically combine infrastructure networks and
MANETs to leverage each other. Specifically,
infrastructure networks improve the scalability of
MANETs, while MANETs automatically establish
self organizing networks, extending the coverage of
the infrastructure networks. Hybrid wireless network
have the characteristics of high mobility and
fluctuating bandwidth. So guarantying the QoS still
remains an open question.

II. EXISTING SYSTEM
As wireless communication gains popularity,
real-time transmission with for wireless applications.
At the same time, a wireless hybrid network that
integrates a mobile wireless ad hoc network
(MANET) and a wireless infrastructure network has
been stringent Quality of Service (QoS) requirements
proven to be a better alternative for the next
generation wireless networks. By directly adopting
resource reservation-based QoS routing for
MANETs, hybrids networks inherit invalid
reservation and race condition problems in MANETs.
How to guarantee the QoS in hybrid networks
remains an open problem. In this paper, we propose a
QoS-Oriented Distributed routing protocol (QOD) to
enhance the QoS support capability of hybrid
networks. Taking advantage of fewer transmission
hops and any cast transmission features of the hybrid
networks, QOD transforms the packet routing
problem to a resource scheduling problem. QOD
incorporates five algorithms: (1) a QoS-guaranteed
7
th

neighbor selection algorithm to meet the transmission
delay requirement, (2) a distributed packet scheduling
algorithm to further reduce transmission delay,(3) a
mobility-based segment resizing algorithm that
adaptively adjusts segment size according to node
mobility in order to reduce transmission time, (4) a
traffic redundant elimination algorithm to increase
the transmission throughput, and (5) a data
redundancy elimination based transmission algorithm
to eliminate the redundant data to further improve the
transmission QoS. Analytical and simulation results
based on the random way-point model and the real
human mobility model show that QOD can provide
high QoS performance in terms of overhead,
transmission delay, mobility resilience and
scalability.

III. PROPOSED SYSTEM

As wireless communications gains popularity,
real time transmission with stringent quality of
service requirements for wireless applications. At the
same time, a wireless hybrid network that integrates a
mobile wireless adhoc network and a wireless
infrastructure network has been proven to be a better
alternative for the next generation wireless network.
protocol enhance the Quality of Service capability of
hybrid networks. In this paper improve the quality of
protocol and develop a protocol Enhanced Quality of
Enhanced Quality Of service Oriented Distributed
routing protocol is mobility resilient than Quality of
routing protocol improves the throughput and
routing protocol can provide high Quality of Service
performance in terms of overhead, transmission delay
and mobility resilient. This also increase the energy
efficiency of Quality of service Oriented Distributed
routing protocol and avoid the energy harvesting
problem by using Enhanced Quality of service
Oriented Distributed routing protocol. Hence it leads
to develop a high efficiency hybrid network using
routing protocol.
3.1 Hybrid Wireless Network

A wireless hybrid network that integrates a
mobile wireless ad hoc network (MANET) and a
wireless infrastructure network. Multi-hop cellular
networks (also called hybrid networks) appear to be a
promising combination of the dynamics of mobile ad
hoc networks and the reliability of infrastructure
wireless networks. These hybrid networks oers
several advantages for users as well as operators. The
network topology can be dynamically adapted to the
respective needs reducing installation costs for the
provider, the overall coverage area can be extended
and nodes can reduce their energy consumption for
transmitting packets due to shorter distances.
However, several weaknesses known from mobile ad
hoc networks persist. In the context of hybrid
networks new possibilities to deal with these
weaknesses become available. Besides the security
and routing issues the cooperation among nodes is of
great importance.

3.2 Introduction To Manet
Hybrid network is the integration of Manet and
wireless infrastructure network. This is the best
alternative for next generation network. Ad-hoc
networks are mobile networks that operate in the
absence of any fixed infrastructure, employing peer-
to-peer communication to establish network
connectivity. These networks have a wide range of
applications such as disaster relief and field
operations, war front activities, and communication
between automobiles activities and it is a self starting
dynamic network comprising of mobile nodes, where
each and every participation node voluntarily
transmit the packets destined to some remote node
using wireless (radio signal) transmission. An ad hoc
network doesnt have any centralized arbitrator or
server. In MANET each and every mobile node is
assumed to be moving with more or less relative
speed in arbitrary direction. Because of that there is
no long term guaranteed path from any one node to
other node. MANET have very enterprising use in
emergency scenarios like military operations &
disaster relief operation where there is need of
communication network immediately following some
major event, or some temporary requirement like
conference & seminar at new place where there is no
earlier network infrastructure exist and need
alternative solution.
Ad hoc network [5][8] is a network where there
is no existence of wireless infrastructure for
networking, Instead each node communicates with
each other using their sole transmitter receiver only.
In this kind of network each and every node does
participate voluntarily in transit packet that flow to
and from different nodes. Each node do follow same
7
th

routing algorithm to route different packets. Thus this
kind of network have limited homogenous feature.
There are not many wireless products that follow this
proposed technology.

.3.3wireless Infrastructure Network

Network with existing infrastructure is a network
where exists a wireless access point or earlier
wireless hardware support for each node to connect
to networks. Here nodes do not participate in any
kind of transit services. They communicate to access
points to send & receive packets from other nodes. In
this kind of network different access point can follow
different wireless protocol like 802.11 b or 802.11g
and still can communicate with each other. There
exist several wireless products based on this kind of
technology .Most wireless networks are based on the
IEEE 802.11 standards [4]. A basic wireless
network consists of multiple stations communicating
with radios that broadcast in either the 2.4GHz or
5GHz band, though this varies according to the locale
and is also changing to enable communication in the
2.3GHz and 4.9GHz ranges.802.11 networks are
organized in two ways. In infrastructure mode, one
station acts as a master with all the other stations
associating to it, the network is known as a BSS, and
the master station is termed an access point (AP). In a
BSS, all communication passes through the AP; even
when one station wants to communicate with another
wireless station, messages must go through the AP.
In the second form of network, there is no master and
stations communicate directly. This form of network
is termed an IBSS and is commonly known as an ad-
hoc network.
3.4 Development Of EQOD From Conventional
Protocols
- A differentiated quality of service oriented
multimedia multicast protocol
In Modern Multimedia communication, there are
some flows that have constraints different from
others and the required QoS for each flow is not the
same. Furthermore, in MC communications, all the
users do not want or are not able to receive the same
QoS. These constraints imply that new
communication mechanisms have to take into
account the user requirements in order to provide an
ad hoc service to each user and to avoid wasting the
network resources. This dissertation proposes a new
differentiated QoS multicast architecture, based on
client/server proxies, called M-FPTP, which relays
many MC LANs by single partially reliable links.
This architecture provides a different QoS to each
LAN depending on the users requirements. For doing
so, it is also provided a network model called
Hierarchies Graph (HG) which represents at the same
time the network performances and the users QoS
constraints. Nevertheless, the application of standard
tree creation methods on an HG can lead to source
overloading problems. It is then proposed a new
algorithm called Degree-Bounded Shortest-Path-Tree
(DgB-SPT) which solves this problem. However, the
deployment of such a service needs a new protocol in
order to collect users requirements and correctly
deploy the proxies. This protocol is called Simple
Session Protocol for QoS MC (SSP-QoM). The
proposed solutions have been modeled, verified,
validated and tested by using UML 2.0 and TAU G2
CASE tool.
- QoS oriented Opportunistic Routing protocol
for Wireless Sensor Networks
In this propose QOR, short for QoS oriented
Opportunistic Routing protocol for data collection in
Wireless Sensor Networks. Unlike classic routing
schemes, QOR takes advantage of opportunistic links
to provide faster and more reliable transmissions. Our
contribution is threefold. First, propose a joint routing
structure and addressing scheme that allows
identifying a limited set of nodes than can become
opportunistic relayers between a source sensor and
the sink. Second, define an original cascaded
acknowledgement mechanism that brings reliable
acknowledgment and replication-free forwarding to
the opportunistic communication scheme. Finally, the
performance evaluation assesses that QOR efficiently
uses opportunistic links to provide reliable and
replication-free data delivery.
- Quality-of-service-oriented protocol for
supporting multimedia services over a
stratospheric platform communication
network
A quality-of-service-oriented medium access
protocol "MAC" protocol is suggested for delivering
multimedia services through a stratospheric
aeronautical platform wireless communication
system. The invented protocol exploits the statistical
multiplexing of asynchronous transfer mode "ATM"
technology. Combing the reservation- and
contention-based access schemes in a single protocol
allows the platform communication system at an
altitude of 20 km to guarantee the service quality
requirements for the diverse services. Exploiting the
flexibility of the protocol as well as the low
encountered propagation delay of the wireless link
permit constant bit rate "CBR", variable bit rate
7
th

"VBR" and available bit rate "ABR" services to be
efficiently multiplexed without violating quality
constraints. The effects of channel capacity and its
associated limitations on the network performance
are discussed and pragmatic solutions are suggested.
Different service priority schemes are presented and
numerical results are discussed. The obtained results
dictate the wireless ATM platform communication as
a promising means for the next-generation wireless
communication system..
- QoS-Oriented Asynchronous Clustering
Protocol in Wireless Sensor Networks

In this propose a QoS-oriented events-driven
asynchronous clustering protocol, called EEAC
(Energy-Efficient Asynchronous Clustering), which
can deliver traffic in a timely and reliable manner. In
EEAC, clustering starts asynchronously according to
a probability, determined by cluster-heads data
transmission rate and residual energy. EEAC avoids
time synchronization and adopts composite formula
to elect cluster heads. Simulation results show that
EEAC ensures the real-time transmission of sensitive
data, reduces the packet loss rate, and evenly
distributes nodal energy consumption, thus
prolonging network lifetime.

- Zone Routing Protocol
Zone Routing Protocol or ZRP is a hybrid
routing protocol that uses both proactive and reactive
routing protocols when sending information over the
network. ZRP was designed to speed up delivery and
reduce processing overhead. It does this by selecting
the most efficient type of protocol to use throughout
the route.
- Quality Of Service Oriented Distributed
Routing Protocol
The above protocols cant provide the quality of
service in the manet.It cause the invalid reservation
and race condition problem. So propose a QoS-
Oriented Distributed routing protocol (QOD) to
enhance the QoS support capability of hybrid
networks. Taking advantage of fewer transmission
hops and any cast transmission features of the hybrid
networks, QOD transforms the packet routing
problem to a resource scheduling problem. QOD
incorporates five algorithms: (1)A QoS-guaranteed
neighbour selection algorithm to meet the
transmission delay requirement.(2)A distributed
packet scheduling algorithm to further reduce
transmission delay.(3) A mobility-based segment
resizing algorithm that adaptively adjusts segment
size according to node mobility in order to reduce
transmission time.(4) A traffic redundant elimination
algorithm to increase the transmission throughput.(5)
A data redundancy elimination based transmission
algorithm to eliminate the redundant data to
further improve the transmission QoS.

- Enhanced Qod
In this improve the quality of QOD and develop
a protocol EQOD.The EQOD is mobility resilient
than QOD.The EQOD improves the throughput and
EQOD can provide high QOS performance in terms
of overhead, transmission delay and mobility
resilient. In this also analyze the energy from the
EQOD protocol and avoid the energy harvesting
problem. In QOD the mobility is about 140km/hr, but
in EQOD the mobility is about 72km/hr.When the
mobility of the EQOD decreased, it decreases the
channel break down and improve the quality.

REFERENCES

[1] H. Wu and X. Jia. QoS multicast routing by
using multiple paths/trees in wireless ad hoc
networks. Ad hoc Networks, 2009.

[2] H. Luo, R. Ramjeey, P. Sinhaz, L. Liy, and S.
Lu. UCAN: A unified cell and ad-hoc network
architecture. In Proc. of MOBICOM, 2003.

[3] P. K. Mckinley, H. Xu, A. Esfahanian, and L. M.
Ni. Unicast-based multicast communication in
wormhole-routed direct networks. IEEE TPDS, 1992.

[4] H. Wu, C. Qiao, S. De, and O. Tonguz.
Integrated cell and ad hoc relaying systems: iCAR. J-
SAC, 2001.

[5] J. Zhou and Y. R. Yang. PAR CelS: Pervasive
ad-hoc relaying for cell systems. In Proc. of Med-
Hoc-Net, 2002.

[6] R. Braden, D. Clark, and S. Shenker. RFC1633:
Integrated services in the internet architecture: an
overview. IETF, 1994.

[7] E. Crawley, R. Nair, B. rajagopalan, and H.
Sandick. RFC2205: Resource reservation protocol
RSVP. IETF, 1998.

7
th

[8] I. Jawhar and J. Wu. Quality of service routing in
mobile ad hoc networks. Springer, Network Theory
and Appli., 2004.

[9] T. Reddy, I. Karthigeyan, B. Manoj, and C.
Murthy. Quality of service provisioning in ad hoc
wireless networks: a survey of
issues and solutions. Ad hoc Networks, 2006.
[10] X. Du. QoS routing based on multi-class nodes
for mobile ad hoc networks. Ad hoc Networks, 2004.

[11] S. Jiang, Y. Liu, Y. Jiang, and Q. Yin.
Provisioning of adaptability to variable topologies for
routing schemes in MANETs. IEEE JSAC,
2004.

[12] M. Conti, E. Gregori, and G. Maselli. Reliable
and efficient forwarding in ad hoc networks. Ad hoc
Networks, 2004.

[13] G. Chakrabarti and S. Kulkarni. Load
banlancing and resource reservation in mobile ad hoc
networks. Ad hoc Networks, 2006.

[14] A. Argyriou and V. Madisetti. Using a new
protocol to enhance path reliability and realize load
balancing in mobile ad hoc networks. Ad hoc
Networks, 2006.

[15] C. Shen and S. Rajagopalan. Protocol-
independent multicast packet delivery improvement
service for mobile ad hoc networks. Ad Hoc
Networks, 2007.

[16] C. E. Perkin, E. M. Royer, and S. R. Quality of
service in ad hoc on-demand distance vector routing.
IETF, 2001.

[17] Z. Shen and J. P. Thomas. Security and QoS
self-optimization in mobile ad hoc networks. IEEE
TMC, 2008.

7
th

Detecting Misbehavior Nodes in
Mobile Ad-Hoc Network Using Intrusion Detection System.

LAVANYA.K M.S.VINU
PG Student, Dept. of CSE, Assistant Professor, Dept of CSE,
Sri Eshwar College Of Engineering, Sri Eshwar College Of Engineering,
Coimbatore. Coimbatore.

lavanya030891@gmail.com vinuja@gmail.com

Abstract:
On the contrary to traditional network
architecture, MANET does not require a fixed
network infrastructure; every single node works as
both a transmitter and a receiver. Nodes
communicate directly with each other when they
are both within the same communication range.
Otherwise, they rely on their neighbors to relay
messages. The self-configuring ability of nodes in
MANET made it popular among critical mission
applications like military use or emergency
recovery. However, the open medium and wide
distribution of nodes make MANET vulnerable to
malicious attackers. In this case, it is crucial to
develop efficient intrusion-detection mechanisms
to protect MANET from attacks. With the
improvements of the technology and cut in
hardware costs, we are witnessing a current trend
of expanding MANETs into industrial
applications. To adjust to such trend, we strongly
believe that it is vital to address its potential
security issues. In this paper, we propose and
implement a new intrusion-detection system
named Enhanced Adaptive ACKnowledgment
(EAACK) specially designed for MANETs.
Compared to contemporary approaches, EAACK
demonstrates higher malicious-behavior-detection
rates in certain circumstances while does not
greatly affect the network performances.

Keywords ACK, S-ACK, IDS, Security issue,
MANET, EAACK.

I. INTRODUCTION:
Mobile Ad-hoc network is a self configuring
infrastructure which nodes act as a both sender and
receiver. In this there is no centralized server for
communicating because this network is independent
infrastructure[1][2]. Nodes transmit the packets within
the range, but it does not transmit the packets when
the receiver beyond the limited range. This leads to
loss of packets and both the node are reaches within
range then it retransmits the packets to respective
receiver or node. When the two nodes are sending the
packets to another node at the same time then it leads
to Packet collision[6]. Then the receiver cannot send
the acknowledgement within the time. Then it sends
the false acknowledgement to sender and it
retransmits the packets.

Here we introduce Intrusion Detection
System for detecting the vulnerabilities and malicious
attack. Intrusion is used to compromise the security,
confidentiality etc., within the nodes and leads to loss
of packets and sends the false acknowledgement,
negative acknowledgement and vulnerability. The
major drawback in this independent infrastructure is
the absence of the centralized server for
communication. If the nodes transmit the packets
bandwidth should be higher but in this network there
is only limited bandwidth. This Intrusion Detection
System used to detect the malicious attack and the
selfish nodes in the MANET. The designing efficient
IDS are used to improve the performance of the
network. Detecting the intrusion is very difficult in
MANET; this can differentiate anomalous activities in
network. There different solutions for detecting the
malicious attack.
In MANET the packets are sent by the sender
and then acknowledgement is received by the sender.
When the packets are send to receiver with the
intermediate node and it sends the TWO
acknowledgements. If the connections are lost then
the receiver does not send the acknowledgement, then
it doesnt knows what happen to the packets. For this,
receiver sends the false acknowledgement then the
sender retransmits the packets[8][10]. Another
solution for reliability is Flooding-based route
discovery in MANETs. This is to set up the route with
reliability between sender and receiver. But this
approach may cause a serious conflict in information
transfer between adjacent nodes and a considerable
amount of control packets. The transmitting of
information between nodes is made secured by
Intrusion detection system (IDS)[16]. Not only the
packets and also have solution for other security
issues in MANET.
MANET is capable of creating a self-
configuring and self-maintaining network without the
help of a centralized infrastructure, which is often
infeasible in critical mission applications like military
conflict or emergency recovery[5]. Minimal
configuration and quick deployment make MANET
ready to be used in emergency circumstances where
an infrastructure is unavailable or unfeasible to install
7
th

in scenarios like natural or human-induced disasters,
military conflicts, and medical emergency situations.
II. SUSCEPTIBILITY IN MANET:
In MANET the vulnerabilities are nature which
is harmful to nodes that are transmitting the packets.
There is no stable infrastructure leads to lack of
boundary and the packets are lost[22]. There are
various link attacks that can expose the mobile ad hoc
network, which make harder for the nodes in the
network to oppose the attacks. These attacks include
message relay, eavesdropping, loss of secret data,
denial of service.

2.1 I DS Architecture:

Fig 2.2.1 IDS Architecture

In the IDS architecture there are many
numbers of independent nodes which is connected to
the Detection system for detecting the malicious
attack. Lack of reliability between the nodes leads to
inconsistent communication of the nodes. This
lacking is due to limited resources for the wireless
nodes. Frequent changes in topology due to mobility
of nodes and independent infrastructure, this affect the
routing information between the nodes[18]. Because
of changing topology each node should be incorporate
with their neighbor node and to avert from the attacks
which may act as liability in the routing protocol. This
architecture is used for IDS to detect the vulnerability
which separate the anomalous activities in the
network.
MANET does not have the centralized
system, named server, which leads to some
susceptible attacks. In large network it is difficult to
identify the traffic and to monitor the system in the
network. These attacks are not easy to detect because
they often change their pattern within the short time.
But this can be found out in system view because
these attacks perform various misbehaviors which
make system failure. But it can be overcome by
considering the trusted nodes. These nodes should be
co-operating with network where security is not
assumed to all nodes[3][9]. For transmitting the
packets to all nodes, cooperative algorithm is used for
all nodes and infrastructure. The network is
decentralized that breaks the cooperative algorithm
and perform the attacks and the vulnerability in
networks.

2.2 Limited Power Supply:
In wired network, no need to consider power
supply because the power is supplied by the wired one
or by their outlets. But in wireless network there is
limited power supply to the nodes because of the
mobility of the nodes. The nodes are exhausted when
it transmit the extra packets which is meaningless one.
This may leads to the loss of packets and the power
supply is limited which cannot be used for reliable
transmission[16][10]. The nodes are behave like a
selfish manner to transmit the packets which it doesnt
help the other nodes in network. This may because the
network vulnerable to the nodes which attacks affects
the other nodes in the network. This also uses the
clustering scheme to prevent the attacks and the
vulnerable of the nodes. Then the monitoring node is
assigned for the cluster scheme which is uses the
intrusion detection technique which can be considered
as the trusted node and no trusted node. These nodes
should be trusted for the transmitting the packets and
this can prevent from the vulnerable malicious attack.

III. PRECAUTIONS FOR ISSUES
For the above vulnerabilities and the attacks
there are much more schemes introduced for the
prevention and detection of these attacks. These
schemes are used to protect the nodes from the
vulnerabilities and from the attacks[17]. This
introduces the Intrusion Detection System for the
nodes to be prevented from the vulnerabilities.

3.1 I DS Techniques in MANET:
Intrusion is the malicious attack that cannot
be detected easily, this can detected by the Intrusion
Detection technique. It detects the unwanted attacks
nodes and the vulnerabilities. In this technique the
IDS Agent is used for the cooperation of the nodes
which compromises the security. This agent consists
of four modules. These modules ensure that detection
of vulnerabilities and the intrusion attacks. In the
figure, they represent the four modules which
allocating a responsibility to each module. In the first
module it checks the data gathered and audit will be
done by the different wealth. Next module is to
scrutinize the local data and detect any vulnerabilities
presence. Third module works with the agent to
7
th

discover any proof that malicious activities is present.
Last module deals with the response that the
malevolent attack is confirmed. Using IDS is to
prevent the malicious attack and mechanism used is
Watchdog, TWOACK, AACK.

Fig 3.1.1 IDS Agents

3.2 I D Response Technique:
Intrusion response technique is composed of
several layers in the detection system for gaining the
better performance even though it is affected by the
vulnerabilities. The layers considered as the MAC
protocol and the application layer[14]. These layer are
used in this technique to integrated the attacks are
easily find in the higher layer such as application
layer than in the lower layer like MAC protocol.
Other than the power supply and the selfish manner
there is another problem is that overhead. The
performance will be lower than the expected one due
to the overhead, so multi layer ID technique is used.
This IDS agent should be present in layer of each
node so that vulnerable can be detected more
efficiently.

3.3 Cluster I D Technique:
In the cluster based intrusion detection
technique, the nodes which are in network should be
cooperative to the other node and network. These may
be down due to the battery power and the selfish
manner of the node in the network[11]. This
technique forms the number of group or cluster in
which a node should at least a member of one group.
In this there should be a one cluster head to monitor
the other nodes and these nodes should be within the
radio range[24][21]. Node is selected as head but the
other should be within the 1-hop vicinity. The election
process is used for the head cluster and this will be
select the other nodes to produce better performance
and with high efficiency and the probability of the
nodes should be equal.

IV. EAACK- SECURE IDS
In MANET, the nodes are connected as the
bi-directional link which acts as both sender and
receiver. When the node sends the message in the
form of packets, it sends back the acknowledgement
but cannot find about the forged acknowledgement.
For this EAACK approach is used MRA scheme for
detecting the forged acknowledgment and report
about the packets that send to other nodes. There are
more schemes like Watchdog, TWOACK and AACK.
These schemes have drawbacks like packet collision,
loss of packets, false report of packets, limited power,
collusion etc. For this a new IDS named EAACK
which packets send the report whether it is a false
report and the forged acknowledgement.

4.1Watchdog
Watchdog that aims to improve the
throughput of network withthe presence of malicious
nodes. In fact, the Watchdog scheme is consisted of
two parts, namely, Watchdog and Pathrater.
Watchdog serves as an IDS for MANETs. It is
responsible for detecting malicious node misbehaviors
in the network[19]. Watchdog detects malicious
misbehaviors by promiscuously listening to its next
hops transmission. If a Watchdog node overhears that
its next node fails to forward the packet within a
certain period of time, it increases its failure counter.
Whenever a nodes failure counter exceeds a
predefined threshold, the Watchdog node reports it as
misbehaving[20]. In this case, the Pathrater
cooperates with the routing protocols to avoid the
reported nodes in future transmission.
Many following research studies and implementations
have proved that the Watchdog scheme is efficient.
Furthermore, compared to some other schemes,
Watchdog is capable of detecting malicious nodes
rather than links. These advantages have made the
Watchdog scheme a popular choice in the field[23].
Many MANET IDSs are either based on or developed
as an improvement to the Watchdog scheme fails to
detect malicious misbehaviors with the presence of
the following: 1) ambiguous collisions; 2) receiver
collisions; 3) limited transmission power; 4) false
misbehavior report; 5) collusion; and 6) partial
dropping.

4.2 TWO ACK
This secure IDS is used to detect the false
misbehavior by the TWO ACK. The solutions for
secure IDS given by S-ACK, MRA (Misbehavior
Report Authentication) and ACK.

7
th

Fig 4.2.1 TWOACK

In this TWOACK, there are 3 nodes which node A
sends the packet to node C which is an intermediate
node B. When A sends the packet to B which in turn
sends to node C and it responds with the
acknowledgement from C to B then from B to A.
when the acknowledgement is not received then the
sender believes that packet is loss due to network
connection and it retransmits the packet to
intermediate node.

4.3 AACK
AACK(Adaptive ACKnowledgement) is an
acknowledgment-based network layer scheme which
can be considered as a combination of a scheme
called TACK (identical to TWOACK) and an end-to-
end acknowledgment scheme called ACKnowledge
(ACK). Compared to TWOACK, AACK significantly
reduced network overhead while still capable of
maintaining or even surpassing the same network
throughput. The end-to-end acknowledgment scheme
in ACK is shown in Fig. 4.3.1 In the ACK scheme
shown in Fig. 4.3.1, the source node S sends out
Packet 1 without any overhead except 2 b of flag
indicating the packet type. All the intermediate nodes
simply forward this packet. When the destination
node D receives Packet 1, it is required to send back
an ACK acknowledgment packet to the source node S
along the reverse order of the same route. Within a
predefined time period, if the source node S receives
this ACK acknowledgment packet, then the packet
transmission from node S to node D is successful.

Fig 4.3.1 AACK Scheme

Otherwise, the source node S will switch to TACK
scheme by sending out a TACK packet. The concept
of adopting a hybrid scheme in AACK greatly reduces
the network overhead, but both TWOACK and
AACK still suffer from the problem that they fail to
detect malicious nodes with the presence of false
misbehaviour report and forged acknowledgment
packets. In fact, many of the existing IDSs in
MANETs adopt
an acknowledgment-based scheme, including
TWOACK and AACK. The functions of such
detection schemes all largely depend on the
acknowledgment packets. Hence, it is crucial to
guarantee that the acknowledgment packets are valid
and authentic.

V. CONCLUSION AND FUTURE
ENHANCEMENT
In this paper we discuss about the security
issues in MANET and it can be prevent using the
Intrusion Detection System. Using IDS there are
many solution are given but there are drawbacks in
reply with the acknowledgement that gives report. But
this report never specifies the details of packets
whether it is losses the packet or collision, or due to
mobility of the nodes. Our future enhancement is said
to overcome the report of the packets and say about
the packet and their details. These have their IDS
named EAACK used to detect the forged
acknowledgements using the Digital Signature
Algorithm (DSA) and compare with the RSA
algorithm through Simulation.

REFERENCES

[1] Marco Conti, Body, Personal and Local Ad Hoc
Wireless Networks, in Book The Handbook of Ad Hoc
Wireless Networks (Chapter 1), CRC Press LLC,
2003.
[2] M. Weiser, the Computer for the Twenty-First
Century, Scientific American, September 1991.
7
th

[3] M.S. Corson, J.P. Maker, and J.H. Cernicione,
Internet-based Mobile Ad Hoc Networking, IEEE
Internet Computing, pages 6370, July-August 1999.
[4] Amitabh Mishra and Ketan M. Nadkarni, Security
in Wireless Ad Hoc Networks, in Book The
Handbook of Ad Hoc Wireless Networks (Chapter
30), CRC Press LLC, 2003.
[5] Lidong Zhou and Zygmunt J. Hass, Securing Ad
Hoc Networks, IEEE Networks Special Issue on
Network Security, November/December 1999.
[6] Y. Hu, A. Perrig and D. Johnson, Wormhole
Attacks in Wireless Networks, IEEE Journal on
Selected Areas in Communications, Vol. 24, No. 2,
February 2006.
[7] A. Perrig, R. Canetti, J. D. Tygar and D. Song,
Efficient Authentication and Signature of Multicast
Streams over Lossy Channels, In Proceedings of the
IEEE Symposium on Research in Security and
Privacy, pages 5673, May 2000.
[8] Jim Parker, Discussion Record for the 1st
MANET Reading Group Meeting,
http://logos.cs.umbc.edu/wiki/eb/index.php/February_
10%2C_2006 (Authorization required).
[9] Architecture for reliable server pooling [Online].
Available: www.ietf.org/ids.by.wg/rserpool.html
[10] M. Fecko, U. Kozat, S. Samtani, M. Uyar, and I.
Hkelek, Architecture and applications of dynamic
survivable resource pooling in battlefield networks,
in Proc. SPIE, vol. 5441, Bellingham, WA, 2004, pp.
204214.
[11] U. C. Kozat and L. Tassiulas, Service discovery
in mobile ad hoc networks: An overall perspective on
architectural choices and network layer support
issues, Ad Hoc Netw., vol. 2, no. 1, pp. 2344, Jan.
2004.
[12] G. Di Crescenzo, G. R. Arce, and R. Ge,
Threshold cryptography for mobile ad hoc
networks, in Lecture Notes in Computer Science.
NewYork: Springer-Verlag, 2004, vol. 3352, Proc.
Security in Commun., pp. 91104.
[13] H. Yang, H. Luo, F. Ye, S. Lu, and L. Zhang,
Security in mobile ad hoc networks: Challenges and
solutions, IEEE Wireless Commun., vol. 11, no. 1,
pp. 3847, Feb. 2004.
[14] Y.-C. Hu and A. Perrig, A survey of secure
wireless ad hoc routing, IEEE Security and Privacy,
pp. 2839, May/Jun. 2004.
[15] B. C. Neuman and T. Tso, Kerberos: An
authentication service for computer networks, IEEE
Commun. Mag., vol. 32, no. 9, pp. 3338, Sep. 1994.
[16] L. Eschenauer and V. D. Gligor, A key-
management scheme for distributed sensor networks,
in Proc. ACM Comput. Commun. Security,
Washington, DC, Nov. 2002, pp. 4147.
[17] W. Diffie and M. E. Hellman, New directions in
cryptography, IEEE Trans. Inf. Theory, vol. IT-22,
no. 6, pp. 644654, Nov. 1976.
[18] L. Lamport, R. Shostak, and M. Pease, The
Byzantine generals problem, ACM Trans. Program.
Languages Syst., vol. 4, no. 3, pp. 382401, Jul. 1982.
[19] A. Boldyreva, Efficient threshold signatures,
multisignature and blind signature schemes based on
the GapDiffieHellman-group signature scheme, in
Lecture Notes in Computer Science. News York:
Springer- Verlag, 2003, vol. 2567, Proc. Public-Key
Cryptography, pp. 3146.
[20] D. Boneh, B. Lynn, and H. Shacham, Short
signatures from the Weil Pairing, in Lecture Notes in
Computer Science. News York: Springer- Verlag,
2001, vol. 2248, Proc. Advances in Cryptology
Asiacrypt, pp. 514532.
[21] J. Katz and M. Yung, Threshold cryptosystems
based on factoring, in Lecture Notes in Computer
Science. News York: Springer-Verlag, 2002, vol.
2501, Proc. Advances in CryptologyAsiacrypt, pp.
192205.
[22] T. Pedersen, A threshold cryptosystem without
a trusted party, in Lecture Notes in Computer
Science. News York: Springer-Verlag, 1991, vol. 547,
Proc. Advances in CryptologyEurocrypt, pp. 522
526.
[23] Y. Zhang and W. Lee, Intrusion Detection in
Wireless Ad-Hoc Networks, Proc. ACM MobiCom
00, Aug. 2000.
[24] A. Herzberg, S. Jarecki, H. Krawczyk, and M.
Yung, Proactive Secret Sharing or: How to Cope
with Perpetual Leakage, Proc. CRYPTO 95, pp.
339-352, 1995.
7
th

ENERGY EFFICIENT TARGET TRACKING TO MAXIMIZE
THE COVERAGE FOR WSN USING SLEEP
SCHEDULING MECHANISM
by
M.PARAMASIVAM
ABSTRACT
In the development of various large-scale sensor systems, a particularly challenging problem is how
to dynamically organize the sensors into a wireless communication network and route sensed information
from the field sensors to a target system. The prime motivation of our work is to balance the inherent trade-
off between the resource consumption and the accuracy of the target tracking in wireless sensor networks.
Toward this objective, the study goes through a new energy-efficient dynamic optimization-based sleep
scheduling and target prediction technique for large-scale sensor networks. We present a probability-based
prediction and optimization-based sleep scheduling protocol (PPOSS) to improve energy efficiency of
proactive wake up. A cluster-based scheme is exploited for optimization-based sleep scheduling. At every
sampling instant, only one cluster of sensors that located in the proximity of the target is activated, whereas
the other sensors are inactive. To activate the most appropriate cluster, we propose a non myopic rule, which
is based on not only the target state prediction but also its future tendency. Finally, the effectiveness of the
proposed approach is evaluated and compared with the state-of-the-art protocols in terms of tracking
accuracy inter node communication, and computation complexity.
7
th

A.R.Vidya
#1
, R.Sridhar
*2
Centre for Advanced Research, Computer Science and Engineering, Regional Centre
Anna University Coimbatore, Coimbatore Dt, Tamilnadu, India
1
vidyaradhakrishan@gmail.com

2
besridharcse@gmail.com

Abstract In the Wireless Sensor Networks
(WSN) a sensor node contains number of sensors that
cooperatively monitors an environment. Each sensor
nodes consists of sensor, memory, battery and
communicating device. The sensor senses the
environment and the sensor node sends those data to
the base station or sink node. Wireless Sensor
Networks are used to maximize the lifetime of the
network. Clustering has proven to be an effective
approach for organizing the network into a connected
hierarchy. Packet dropping and modification are
common attacks that can disrupt communication of
the cluster in the Wireless Sensor Networks. In this
article, we highlight the challenges of clustering in
Wireless Sensor Networks, identifying adverse disrupt
and destroying the particular nodes. Node
categorization and hybrid ranking methods are
implemented for the identification of bad nodes in the
sink. The energy of the cluster head is maintained by
selecting the cluster head simultaneously. The nodes
are ranked based on the probability of being bad
based on the dropping ratio and are deleted. An
extensive analysis and simulations have been
conducted to verify the effectiveness and efficiency of
the scheme.

Keywords Cluster Head, Wireless Sensor
Networks, Cluster Nodes, Distributed Cluster, Ad hoc
On Demand.
INTRODUCTION
Wireless Sensor Networks have emerged as
research areas with an overwhelming effect on
practical application development. They permit fine
grain observation of the ambient environment at an
economical cost much lower than currently possible.
In hostile environments where human participation
may be too dangerous in sensor network which may
provide a robust service. Sensor networks are
designed to transmit data from an array of sensor
nodes to a data repository on a server. The advances
in the integration of MEMS, microprocessor and
wireless communication technology have been
enabled the deployment of large scale. WSN has
potential to design many new applications for
handling emergency, military and disaster relief
operations that requires real time information for
efficient coordination and planning.
Sensors are devices that produce a
measurable response to a change in a physical
condition like temperature, humidity, pressure etc.
WSNs may consist of many different types of sensor
such as seismic, magnetic, thermal, visual, infrared,
and acoustic and radar capable to monitor a wide
variety of ambient conditions. Through each
individual sensor may have severe resource constraint
in terms of energy, memory, communication and
computation capabilities; large number of them may
collectively monitor the physical world and process
the information on the fly environment.
A WSN is different from other popular
wireless networks like cellular network, WLAN and
Bluetooth in many ways. Compared to other wireless
networks, a WSN has much more nodes in a network,
distance between the neighbouring nodes is much
shorter and application data rate is much lower also.
Due to these characteristics, power consumption in a
sensor network will be minimized. To keep the cost of
the entire sensor network down, cost of each sensor
needs to be reduced. It is also important to use tiny
sensor nodes. A smaller size makes it easier for a
sensor to be embedded in the environment it is in.
WSNs may also have a lot of redundant data since
multiple sensors can sense similar information. The
sensed data therefore need to be aggregated to
decrease the number of transmission in the network,
reducing bandwidth usage and eliminating
unnecessary energy consumption in both transmission
and reception.
An Efficient Analysis of Intrusion Detection Using Clustering In
WSN
7
th

The main characteristics of a WSN include,
- Power consumption using batteries
or energy harvesting
- Ability to cope with node failure
- Mobility of nodes
- Heterogeneity of nodes
- Scalability to large scale deployment
- Ease of use
In a WSN, sensor nodes monitor the
environment, detect events of interest, produce data
and collaborate in forwarding the data towards a sink,
which could be a gateway, base station, storage node,
or querying user. A sensor network is often deployed
in an unattended and hostile environment to perform
the monitoring and data collection tasks. When it is
deployed in such an environment, it lacks physical
protection and is subject to node compromise. After
compromising one or multiple sensor nodes, an
adversary may launch various attacks to disrupt the
in-network communication. Among these attacks, two
common ones are dropping packets and modifying
packets, i.e., compromised nodes drop or modify the
packets that they are supposed to forward.
The large-scale deployment of wireless
sensor networks (WSNs) and the need for data
aggregation necessitate efficient organization of the
network topology for the purpose of balancing the
load and prolonging the network lifetime. Clustering
has proven to be an effective approach for organizing
the network into a connected hierarchy. In this article,
we highlight the challenges in clustering a WSN,
discuss the design rationale of the different clustering
approaches, and classify the proposed approaches
based on their objectives and design principles. We
further discuss several key issues that affect the
practical deployment of clustering techniques in
sensor network applications.
In order to support data aggregation through
efficient network organization, nodes can be
partitioned into a number of small groups called
clusters. Each cluster has a coordinator, referred to as
a cluster head, and a number of member nodes.
Clustering results in a two-tier hierarchy in which
cluster heads (CHs) form the higher tier while
member nodes form the lower tier. The member nodes
report their data to the respective CHs. Research on
clustering in WSNs has focused on developing
centralized and distributed algorithms to compute
connected dominating sets. The CHs aggregate the
data and send them to the central base through other
CHs. Because CHs often transmit data over longer
distances, they lose more energy compared to member
nodes. The network may be clustered periodically in
order to select energy-abundant nodes to serve as
CHs, thus distributing the load uniformly on all the
nodes. Besides achieving energy efficiency, clustering
reduces channel contention and packet collisions,
resulting in better network throughput under high
load.
H.Chan and A. Perrig., [1] has expect future
sensor networks to consist of hundreds or thousands
of sensor nodes. Each node represents a potential
point of attack, making it impractical to monitor and
protect each individual sensor from either physical or
logical attack. The networks may be dispersed over a
large area, further exposing them to attackers who
capture and reprogram individual sensor nodes.
Attackers can also obtain their own
commodity sensor nodes and induce the network to
accept them as legitimate nodes, or they can claim
multiple identities for an altered node. Once in control
of a few nodes inside the network, the adversary can
then mount a variety of attacksfor example,
falsification of sensor data, extraction of private
sensed information from sensor network readings, and
denial of service. Addressing the problem of sensor
node compromise requires technological solutions.
V. Bhuse, A. Gupta, and L. Lilien, Dpdsn.,
[3] as Denial-of-service (DoS) attacks on wireless
sensor networks (WSNs) can deplete network
resources and energy without much effort on the part
of an adversary. Packet-dropping attacks are one
category of DoS attacks. Lightweight solutions to
detect such attacks on WSNs are needed. Current
techniques for detecting such attacks in ad hoc
networks need to monitor every node in the network.
Once they detect malicious nodes that drop packets, a
new path has to be found that does not include them.
In this paper, we propose a lightweight solution called
DPDSN. It identifies paths that drop packets by using
alternate paths that WSN finds earlier during route
discovery. Responding to a packet-dropping attack
incurs no additional cost because one of the alternate
paths is utilized for the subsequent communication.
DPDSN does not require monitoring individual nodes,
making it feasible for WSNs. We formulate the
probability of success and failure of DPDSN in the
presence of malicious nodes that drop packets. We
compare our approach with existing techniques. Our
analysis found that the overhead of DPDSN is at most
for a two-dimensional grid network on nodes. Our
simulations show that the overhead of DPDSN for a
WSN with 100 nodes is less than 3% of energy
consumed on route discovery when using DSR or
Directed Diffusion routing protocols.
7
th

R. Roman, J. Zhou, and J. Lopez., [5]
Wireless sensor networks (WSNs) are vulnerable to
different types of security threats that can degrade the
performance of the whole network; that might result
in fatal problems like denial of service (DoS) attacks,
routing attacks, Sybil attack etc. Key management
protocols, authentication protocols and secure routing
cannot provide security to WSNs for these types of
attacks. Intrusion detection system (IDS) is a solution
to this problem. It analyses the network by collecting
sufficient amount of data and detects abnormal
behaviour of sensor node(s). IDS based security
mechanisms proposed for other network paradigms
such as ad hoc networks, cannot directly be used in
WSNs. Researchers have proposed various intrusion
detection systems for wireless sensor networks during
the last few years. We classify these approaches into
three categories i.e. purely distributed, purely
centralized and distributed-centralized. In this paper,
we present a survey of these mechanisms. These
schemes are further differentiated in the way they
perform intrusion detection.
S. Banerjee and S. Khuller., [6] of clustering
scheme to create hierarchical control structure in the
multi-hop wireless networks has many constraints. A
cluster is defined as a subset of vertices, whose
induced graph is connected. In addition, a cluster is
required to obey certain constraints that are useful for
management and scalability of the hierarchy. All these
constraints cannot be met simultaneously for general
graphs, but we show how such a clustering can be
obtained for wireless network topologies. Finally, we
present an efficient distributed implementation of our
clustering algorithm for a set of wireless nodes to
create the set of desired clusters.
K.Ioannis, T.Dimitriou, and F.C.Freiling.,
[12] is Denial-of-Message Attack (DoM), where
sensor nodes are deprived of broadcast messages.
While nodes can fail to receive broadcasts due to
benign network failures, here we consider the
possibility that these failures are maliciously induced
by an attacker. A simple approach is for every
broadcast recipient to send an authenticated
acknowledgment for each broadcast message.
However, this approach results in a substantial load
on the network to carry acknowledgments and on the
base station to process them.
ORGANIZATION OF PAPER
In the WSN, it consists of a system
initialization phase and several equal-duration rounds
of intruder identification phases. In the initialization
phase, sensor nodes form a topology which is a
directed acyclic graph (DAG). A routing tree is
extracted from the DAG. In each round, data are
transferred through the routing tree to the sink. Each
packet sender/ forwarder adds a small number of extra
bits to the packet and also encrypts the packet. When
one round finishes, based on the extra bits carried in
the received packets, the sink runs a node
categorization algorithm to identify nodes that must
be bad and nodes that are suspiciously bad.
According to the scheme, a dynamic routing
tree rooted at the sink is first established. When sensor
data is transmitted along the tree structure towards the
sink, each packet sender or forwarder adds a small
number of extra bits, which is called packet marks, to
the packet. The format of the small packet marks is
deliberately designed such that the sink can obtain
very useful information from the marks. Specifically,
based on the packet marks, the sink can figure out the
dropping rate associated with every sensor node, and
then run our proposed node categorization algorithm
to identify nodes that are droppers/ modifiers for sure
or are suspicious droppers/modifiers. As the tree
structure dynamically changes every certain time
interval, behaviours of sensor nodes can be observed
in a large variety of scenarios. As the information of
node behaviours has been accumulated, and the sink
periodically run our proposed heuristic ranking
algorithms to identify most likely bad nodes from
suspiciously bad nodes. This way, most of the bad
nodes can be gradually identified with small false
positive.
As a certain number of rounds have passed,
the sink will have collected information about node
behaviours in different routing topologies. The
information includes which nodes are bad for sure,
which nodes are suspiciously bad, and the nodes
topological relationship. To further identify bad nodes
from the potentially large number of suspiciously bad
nodes, the sink runs heuristic ranking algorithms.
SYSTEM DESIGN
In a WSN, sensor network is often deployed
in an unattended and hostile environment to perform
the monitoring and data collection tasks. When it is
deployed in such an environment, it lacks physical
protection and is subject to node compromise. After
compromising one or multiple sensor nodes, an
adversary may launch various attacks to disrupt the
in-network communication. Among these attacks, two
7
th

common ones are dropping packets and modifying
packets. Attackers can also obtain their own
commodity sensor nodes and induce the network to
accept them as legitimate nodes, or they can claim
multiple identities for an altered node.
Node categorization ensures that the
clustering process starts simultaneously throughout
the network. Lack of categorization may result in a
suboptimal choice of CHs, especially for probabilistic
approaches. However, the clustering process can be
triggered by nodes with faster clocks. This happens
when such nodes start querying their neighbours for
updated information in order to start the clustering
process. Received queries trigger the clustering
process at these neighbours, and these neighbours in
turn trigger their neighbours, and so on.
An important objective of any clustering
technique is network connectivity. For intra-cluster
communication, a cluster member communicates with
its CH either directly. Connectivity in this case is a
result of the success of cluster formation. For inter-
cluster communication, two approaches were adopted
in order to maintain connectivity. In one approach,
nodes on cluster boundaries are used as gateways to
relay data among CHs. This approach is suitable in
networks that use a fixed transmission power.
Network density has to be sufficiently high in order to
ensure that enough gateways are present at the
intersection areas between clusters.
To balance the trade-off, we further propose
the HR method. According to HR, the node with the
highest accused account value is still first chosen as
most likely bad node. After a most likely bad node
has been chosen, the one has the highest accused
account value among the rest is chosen only if the
node has not always been accused together with the
bad nodes that have been identified already. The
accusation account value is considered as an
important criterion in identification, and the
possibility that an innocent node being framed by bad
nodes is also considered by not choosing the nodes
who have always being suspected together with
already-identified bad node.

System Model
EXPERIMENTAL EVALUATION
Distributed clustering protocols achieve their
best performance when sensor nodes are categorized.
Node categorization ensures that the clustering
process starts simultaneously throughout the network.
Lack of categorization may result in a suboptimal
choice of CHs, especially for probabilistic
approaches. However, the clustering process can be
triggered by nodes with faster clocks. This happens
when such nodes start querying their neighbours for
updated information in order to start the clustering
process. Received queries trigger the clustering
process at these neighbours, and these neighbours in
turn trigger their neighbours, and so on. It is not
essential that all the nodes in the network start the
clustering process simultaneously. In fact, it is
sufficient that the process starts in different regions
simultaneously. Probabilistic techniques that use a
number of iterations (e.g., [3, 5]) are therefore less
impacted by the lack of synchronization than single-
iteration techniques.
If a compromised node modifies the packets
that it is supposed to forward, the node can be
detected with the fore described scheme. This is
because modified packets will be detected by the sink
and thus be dropped. This is equivalent to that the
packets are dropped by the modifier; hence, the
packet modifier can be identified as a packet dropper
using the Hybrid Ranking-Based Approach in cluster
nodes.
Sensor Node Deployment
Large number of sensor nodes are randomly
deployed in a two dimensional area. Each sensor node
generates sensory data periodically and all these
nodes collaborate to forward packets containing the
data towards a sink. The sink is located within the
network. The sink is aware of the network topology,
which can be achieved by requiring nodes to report
their neighbouring nodes right after deployment. The
purpose of sensor node deployment is to set up secret
pair wise keys between the sink and every regular
sensor node, to establish the cluster network and
routing algorithm to facilitate packet forwarding from
every sensor node to the network.
Neighbour Distance Calculation

In iterative clustering techniques, a node
waits for a specific event to occur or certain nodes to
decide their role before making a decision. A node
waits for all its neighbours with higher weights to
7
th

decide to be CHs or join existing clusters. Nodes
possessing the highest weights in their one-hop
neighbourhoods are elected as CHs.
Cluster Node Formation

To support data aggregation through efficient
network organization, nodes can be partitioned into a
number of small groups called clusters. Each cluster
has a coordinator, referred to as a cluster head, and a
number of member nodes. Clustering results in a two-
tier hierarchy in which cluster heads (CHs) form the
higher tier while member nodes form the lower tier.
The member nodes report their data to the respective
CHs. The CHs aggregate the data and send them to
the central base through other CHs. Because CHs
often transmit data over longer distances, they lose
more energy compared to member nodes. The
network may be reclustered periodically in order to
select energy-abundant nodes to serve as CHs, thus
distributing the load uniformly on all the nodes

Packet Sending and Forwarding
When a node wants to send out a packet, it
attaches to the packet with a sequence number,
encrypts the packet only with the key shared with the
sink, and then forwards the packet to the cluster head.
When an innocent intermediate node receives a
packet, it attaches a few bits to the packet to mark the
forwarding path of the packet, encrypts the packet,
and then forwards the packet to its parent.
After receiving a packet, the sink decrypts it,
and thus finds out the original sender and the packet
sequence number. The sink tracks the sequence
numbers of received packets for every node, and for
every certain time interval, which we call a round, it
calculates the packet dropping ratio for every node. `
Node Categorization
In this module, to identify nodes those are
droppers/modifiers for sure or are suspicious
droppers/ modifiers. Behaviours of sensor nodes can
be observed in a large variety of scenarios. In every
round, for each sensor node, the sink keeps track of
the number of packets sent from sensor node, the
sequence numbers of these packets, and the number of
flips in the sequence numbers of these packets. In the
end of each round, the sink calculates the dropping
ratio for each sensor node. The dropping ratio in this
round is calculated based on the dropping ratio of
every sensor node and the cluster based algorithm, the
sink identifies the nodes that are droppers for sure and
that are possibly droppers.
Hybrid Ranking Algorithm

The bad node modification is done by HR
method to reduce packet droppers and modifiers. The
suspiciously bad nodes are identified based on the
simultaneous selection of nodes for sending packet.
For each of these scenarios, node categorization
algorithm is applied to identify sensor nodes that are
bad for sure or suspiciously bad. After multiple
rounds, sink further identifies bad nodes from those
that are suspiciously bad by applying several
proposed heuristic methods. The tree used for
forwarding data from sensor nodes to the sink is
dynamically changed from round to round. In other
words, each sensor node may have a different parent
node from round to round. We rank the suspiciously
bad nodes based on their probabilities of being bad,
and identify part of them as most likely bad nodes.
CONCLUSIONS
Here, the simple yet effective scheme to
identify misbehaving forwarders that drop and modify
packets in the cluster network. The sink recover
source of the packet and figure out the dropping ratio
associated with each sensor node. Also the node is
categorized based on the packet received by the
destination. Our packet dropper/modifier
identification scheme is implemented in the ns-2
simulator (version 2.3) to evaluate the effectiveness
and efficiency of this clustering node. We measure the
performance of our scheme from two aspects: the
detection rate, defined as the ratio of successfully
identified bad nodes, and the false positive
probability, defined as the ratio of adverse innocent
nodes over all innocent nodes. Extensive analysis and
simulations have been conducted and verified the
effectiveness of the proposed scheme in various
scenarios. The implementation provides the effective
scheme to identify the misbehaving forwarders. The
sensor nodes deployment can be set from 5 to 50
numbers randomly. It takes more time for finding the
misbehaving nodes, if there is increase in the number
of nodes. In each round the sensor node keeps track of
the packets send. The dropping ratio is calculated
between the source node and the destination node.
REFERENCES
[1] H.Chan and A. Perrig, Security and Privacy
in Sensor Networks, IEEE
Computer,October 2003.
7
th

[2] C. Karlof and D. Wagner, Secure routing in
wireless sensor networks: attacks and
countermeasures, the First IEEE
International Workshop on Sensor Network
Protocols and Applications, pp. 113127,
May 2003.
[3] V. Bhuse, A. Gupta, and L. Lilien, Dpdsn:
Detection of packet-dropping attacks for
wireless sensor networks, In the Trusted
Internet Workshop, International Conference
on High Performance Computing, December
2005.
[4] S. Marti, T. Giuli, K. Lai, and M. Baker,
Mitigating Routing Misbehavior in Mobile
Ad Hoc Networks, ACM MobiCom,
August 2000.
[5] R. Roman, J. Zhou, and J. Lopez, Applying
intrusion detection systems to wireless
sensor networks, Third IEEE Annual
Consumer Communications and Networking
Conference (CCNC), pp. 640644, January
2006.
[6] S. Banerjee and S. Khuller, A Clustering
Scheme for Hierarchical Control in
MultihopWireless Networks, Proc. IEEE
INFOCOM, Apr. 2001, pp. 102837.
[7] OssamaYounis, Marwan Krunz, and
SrinivasanRamasubramanian, University of
ArizonaNode Clustering in Wireless Sensor
Networks, Recent Developments and
Deployment Challenges.
[8] G. Chen and I. Stojmenovic, Clustering and
routing in wireless ad hoc networks, TR-99-
05, SITE,University of Ottawa (June 1999).
[9] A Path-Connected-Cluster Wireless Sensor
Network and its Formation, Addressing, and
Routing Protocols Chia-Hung Tsai and Yu-
Chee Tseng, Fellow, IEEE
[10] C.-T.Cheng, C. K. Tse, and F. C. M. Lau. A
Clustering Algorithm for Wireless Sensor
Networks Based on Social Insect Colonies,
IEEE Sensors Journal, 11(3):711721, 2011.
[11] M. Ali and Z. A. Uzmi. An Energy-
Efficient Node Address Naming Scheme for
Wireless Sensor Networks, In Proc. of IEEE
Intl Networkingand Communications
Conference (INCC), 2004.
[12] K.Ioannis, T.Dimitriou, and F.C.Freiling,
Detection of Denial-of-Message Attacks on
Sensor Network Broadcasts, In Proc. of
IEEE Symp.Security and policy, 2005

7
th

ANALYSING THE PERFORMANCE OF SCHEDULING
ALGORITHMS IN WIMAX NETWORK WITH CBR TRAFFIC
USING QUALNET

by
SURESH.M

ABSTRACT: This Scheduling algorithm can be
implemented using a variety of applications such
as CBR, VBR, UBR and ABR but in this research
we have proposed scheduling algorithms using the
application CBR and the performance of WIMAX
network is evaluated using the different
parameters like throughput, delay, jitter and
delivery ratio. By using this analysis we can find
out best scheduling algorithm in WIMAX with
CBR application.

KEYWORDS: CBR, VBR, ABR, UBR, throughput,
jitter and delay.

I. INTRODUCTION:

The IEEE Standard 802.16-2001 defines the
Wireless MAN air interface specification for wireless
metropolitan area networks (MANs). The completion
of this standard signs the entry of broadband wireless
access as a major new tool in the effort to link end
stations to core telecommunications networks
worldwide. The IEEE 802.16 MAC protocol was
specially designed for the point to multipoint
broadband connection for the wireless connection.
The WIMAX MAC layer supports both the
continuous and burst nature traffics. It can be used in
both point to point (P2P) and the typical WAN type
configurations. The WiMAX supports different
multimedia applications as VoIP, voice conference
and online gaming. The IEEE 802.16 technology
(WiMAX) is an improved option to 3G or wireless
LAN networks for providing ease of access, low cost
and large coverage area. In the WIMAX network
there are two technologies they are IEEE 802.16 this
for broadband wireless connection and another one is
IEEE 802.16e this standard is for mobile broadband.
WiMAX scramblers can selectively scramble control
or management messages with the aim of affecting the
normal operation of the network. Slots of data traffic
belonging to the targeted SSs can be scrambled
selectively, forcing them to retransmit. Noise
jamming and multi-carrier jamming are considered
here for simulation approach connection.
WiMAX operates both in 10GHz-66GHz
(licensed frequency band) as well as 2 GHz-11 GHz
(unlicensed frequency band) for Line of Sight (LOS)
and Non-line of Sight (NLOS) operation respectively.
The WiMAX Network technology is an evolutionary
one as it uses orthogonal frequency division
multiplexing (OFDM) which makes transmission
resist fading and minimizes multipath effect. It
supported services in IEEE 802.16d fixed, nomadic
and portable and for IEEE 802.16e mobile fixed,
nomadic and portable. The coverage area for IEEE
802.16d up to 50km maximum and for mobile
broadband 2-5 km approximately. The WIMAX
implementation broadband connection 256-OFDM
scalable and for mobile broadband OFDMA as mobile
WIMAX

WiMAX provides Quality of Service (QoS)
that supports five different categories of services
namely: Unsolicited grant services (UGS), Real-time
polling services (rtPS), Non- real-time polling service
rate (nrtPS), extended real-time polling service
(ertPS) and Best-Effort services (BE). WiMAX can
be used for wireless networking like the popular Wi-
Fi. WiMAX, a second-generation protocol, allows
higher data rates over longer distances, efficient use
of bandwidth, and avoids interference almost to a
minimum. WiMAX can be termed partially a
successor to the Wi-Fi protocol, which is measured in
feet, and works, over shorter distances. Fixed wireless
is the base concept for the metropolitan area
networking (MAN) given in the 802.16 standard. In
fixed wireless, a backbone of base stations is
connected to a public network.

II. PROTOCOL USED IN THIS PAPER:

DYNAMIC SOURCE ROUTING PROTOCOL:

DSR allows the network to be completely
self-organizing and self-configuring, without the need
for any existing network infrastructure or
administration. DSR has been implemented by
numerous groups, and deployed on several test beds.
Networks using the DSR protocol have been
connected to the Internet. DSR can interoperate with
7
th

Mobile IP, and nodes using Mobile IP and DSR have
seamlessly migrated between WLANs, cellular data
services, and DSR mobile ad hoc networks. The
protocol is composed of the two main mechanisms of
"Route Discovery" and "Route Maintenance", which
work together to allow nodes to discover and maintain
routes to arbitrary destinations in the ad hoc network.
All aspects of the protocol operate entirely on-
demand, allowing the routing packet overhead of DSR
to scale automatically to only that needed to react to
changes in the routes currently in use.

ROUTE DISCOVERY:

Route discovery is the mechanism that is
used by a source node which wishes to send data
packets to a destination node which has no route to it
in its route cache. Using this mechanism the source
node can obtain a source route to the destination. For
route discovery, the source node starts by
broadcasting a route request packet that can be
received by all neighbour nodes within its wireless
transmission range. The route request contains the
address of the destination host, referred to as the
target of the route discovery , the sources address, a
route record field and a unique identification number
In this case no periodic routing updates are needed,
providing substantial savings in network bandwidth
and battery power requirement for all involved. A
general solution to route discovery in Ad hoc
Networks is a technique for extending this to the case
in which source and target may not be within the
range of each other.

ROUTE MAINTENANCE:

Route maintenance is the
mechanism that is used by a source node to detect a
link breakage along its source route to a destination
node. Using this mechanism the source node can
know if it can still use the route or not. When the
source node indicates the existence of a broken link in
the source route, it can use another route or trigger a
new route discovery process. Route maintenance is
used only with active routes. Conventional routing
protocols integrate route discovery with route
maintenance by continuously sending the normal
periodic routing updates. In many wireless networks,
route maintenance can be provided with very little
overhead. Since wireless networks are inherently less
reliable than the wired networks. While route is in
use, the route maintenance procedure monitors the
operation of the route and informs the sender of any
routing errors. Many wireless networks utilize hop by
hop acknowledgement at the data link level in order to
provide early detection and retransmission of lost or
corrupted packets.

III. SIMULATION ENVIRONMENT:

The simulation process of Wimax is implemented
using simulator Qualnet 5.0.2 as per the Simulation
Parameters shown in the Table 1. Performance
metrics used in this paper are End-to-End delay,
Average jitter, Packet delivery ratio, Throughput.
The obtained result is considered for IEEE standard
Wimax. CBR (Constant bit rate) application is used
for in the scenarios. The nodes used in the scenarios
are 25, 50, 75 and 100.

Fig.1- Simulation scenario for Wimax.

TABLE.1 Simulation parameters

Channel Wireless

Propagation Two way ground
model

MAC layer 802.16

Dimension 1000m x1000m

Antenna model Omni directional

Routing protocol DSR

Simulation time 300s

No of nodes 25, 50, 75, 100

IV. PERFORMANCE METRICS:
Throughput:

It is nothing but successful data packets
received at the destination node through the
communication channel.. It is usually measured
through bits per second (b/s). It is nothing but total
number of packets received from source to
destination.

Average jitter:

It is occurring due to the deviation among
the packets it caused some interruption like
congestion, timing drift, etc. It is calculated using the
7
th

formula reception time (I)-reception time (I-1).

Average end-to-end delay:

Total End-to-End delay divided by no of
packets received. It is the no of packets received from
source to the destination. It occurs due to the
congestion.

Packet delivery ratio:

The packet delivery ratio is calculated using
the number of packets received from source to
destination without any drop during the transmission.
The simulation scenario is explained in Fig.1.

V. SIMULATION RESULTS:

Throughput: In Fig-2 DSR protocol by comparing
four scheduling algorithm it is finalized that the
Weighted round robbin gives better performance
compare to others for the Wimax.

Fig.2. Throughput of Wimax using DSR Protocol.

Average jitter: In Fig-2 DSR protocol by comparing
four scheduling algorithm it is finalized that the
Weighted round robin gives better performance
compare to others for the Wimax.

Fig.3. Average jitter of Wimax using DSR protocol.

End to end delay: In Fig-2 DSR protocol by
comparing four scheduling algorithm it is finalized
that the Weighted round robin gives better
performance compare to others for the Wimax,

Fig.4. Average end-to-end delay of Wimax using
DSR protocol.

Packet delivery ratio: In Fig-2 DSR protocol by
comparing four scheduling algorithm it is finalized
that the Weighted round robbin gives better
performance compare to others for the Wimax.

Fig.5. Packet delivery ratio of Wimax using DSR
protocol.
VI. Conclusion:

In this paper we have evaluated standard IEEE 802.16
or Wimax. Our work in this paper is to find out the
best scheduling algorithm by using the following
parameters throughput, average jitter, average end-to-
end delay, packet delivery ratio these parameter is
considering in the DSR protocol. In these scheduling
algorithm we have consider for this paper is Strict
Priority ,

Round Robin , Weighted Fair , Weighted Round
Robin by analysing all these four algorithms the
Weighted round robin gives better out in all
parameters.

References:

[1]Anjali Aggarwal, Universal Quality of
Service (QoS) in the New Public Network
Architecture, IEEE Canadian Review-Fall/Automne
2000.

D. ITU-T Recommendation G.711 Appendix I:
A High Quality Low-Complexity Algorithm for
Packet Loss Concealment

with G.711"

E. Masood Khosroshahy and Vivien Nguyen. A
Study of WiMax QoS Mechanisms.RMOB Project,
Telecom Paris; Apr. 2006; 1-23.
7
th

F. Ara Khil and Seungryoul Maeng. Scheduling
of Multimedia Traffic for Continuous Media in
Packet-Switched Networks. Multimedia Computing
and Networking; Jan. 1997;3020: 29-40.
http://dx.doi.org/10.1117/12.264305.
G. Matthew Andrews and Bell Laboratories.
Probabilistic End-to-End Delay Bounds for Earliest
Deadline First Scheduling. Nineteenth Annual Joint
Conference of the IEEE Computer and
Communications Societies; Mar. 2000; 2: 603-612.
http://dx.doi.org/10.1109/INFCOM.2000.83 2234.
H. Qualnet 5.0.2 Advanced Wireless Model Library,
Scalable Network Technologies, Los Angeles; 2010.
I. B.Skrikar. Packet Scheduling Algorithms to
Support QoS in Networks. Masters Paper, Indian
Institute of Technology; Oct. 1999; 1-71.
J. Y. Cao and V. Li. Scheduling
Algorithms in Broadband Wireless
Networks. Proceedings of The IEEE; Jan. 2001; 89:
76-87. http://dx.doi.org/10.1109/5.904507.
[9] Application Layer Threats to WiMAX
Technology, [Online] [accessed March 2011].
Available from URL
www.freewimaxinfo.com/application-layer-
threat.html.

[10] Bo Han, Weijia Jia, Lidong Lin, Performance
evaluation of scheduling in IEEE 802.16 based
wireless mesh networks. Computer Communications,
Vol. 30, Issue 4. Pp 782-792,
http://dx.doi.org/10.1016/j.comcom.2006.10. 001.
[11] Fen Hou, Lin X. Cai, James She, Pin-Han Ho,
Xuemin (Sherman) Shen, and Junshan Zhang.
Cooperative Multicast Scheduling Scheme for IPTV
Service over IEEE 802.16 Networks. IEEE
International Conference; May 2008; 2566 - 2570.
http://dx.doi.org/10.1109/ICC.2008.486.
[12] Yi-Ting Mai, Chun-Chuan Yang, and Cheng-
Jung Wen. A New CutThrough Mechanism in IEEE
802.16 Mesh Networks. Proceedings of World
Academy of Science, Engineering and Technology;
June 2009; 41: 98-104.

[13] Konark kelaiya. Routing & Scheduling
Algorithm of IEEE 802.16 Mesh Backhaul Network
for Radio Recourse Management (RRM). Mobile
and Pervasive Computing; Oct. 2008; 175-179.
7
th

A Novel Method to Reduce Rebroadcast Redundancy in MANETs

V.Hema M.Mohanapriya

PG Student

Department of Computer Science and Engineering

Sri Krishna College of Engineering and Technology

Coimbatore

hemsinfotech22@gmail.com

Abstract - Routing is one of the challenging issues in
Mobile Ad hoc NETworks (MANETs). Broadcasting is the
fundamental and efficient data dissemination mechanism
for route discovery in reactive routing protocols of Mobile
Ad hoc NETwork (MANET). This causes the problem
called the broadcast storm problem which results in
redundant retransmission and adds to routing overhead.
There are many approaches proposed to solve the
problem; but none of them addresses the problem
effectively. This paper proposes a new mechanism that has
probabilistic rebroadcast based on neighbor coverage for
the routing overhead reduction. This proposed mechanism
will reduces the packet retransmission and thus reduce the
routing overhead. This approach combines the advantages
of probabilistic mechanism and neighbor area coverage
based approach. This new mechanism can improve the
performance of broadcasting in various network scenarios.
This approach is simple and can be implemented in NS-2.

Keywords - Ad Hoc Network, broadcast storm, MANET,
probabilistic rebroadcasting

I. INTRODUCTION

Mobile Ad Hoc Networks (MANETs) consist of nodes that
change position frequently. Each node in a mobile ad hoc
network functions as both a host and a router, and the control
of the network is distributed among the nodes. The network
topology is dynamic, because the connectivity among the
nodes may vary with time due to node departures, new node
arrivals, and also due to movement of nodes. The re-active
routing protocols (or on-demand protocols) [13, 6] starts a
route discovery procedure when needed. When a route from a
source to a destination is needed, route searching procedure is
started. Due to increase in the movement of nodes in mobile
ad hoc networks (MANETs), frequent link breakages occurs
often which results in frequent path failures and needs route

Assistant Professor

Department of Computer Science and Engineering

Sri Krishna College of Engineering and Technology

Coimbatore

discoveries. The conventional reactive routing protocol [13, 6]
uses flooding to find the routes between source and
destination. It simply broadcast the route request packet when
the path is needed. The process continues until it finds the
route to the destination. This broadcasting induces the
redundant retransmission. This further causes overhead in
route discovery. Broadcasting is the basic and fundamental
data dissemination mechanism, in which a mobile node
rebroadcasts the route request packets until it has a route to
the required destination, and this causes the broadcast storm
problem [11]. This paper implements the new broadcasting
technique to reduce the overhead of routing packets. This
technique exploits the neighborhood knowledge using
rebroadcast delay and also obtains the coverage ratio of a
node. The connectivity factor and the coverage ratio is used to
calculate the rebroadcast probability. The connectivity factor
is used to determine the number of nodes that need to receive
the Route Request packet. In order to reduce the overhead, the
periodical Hello packets are not used. Since a node sending
any broadcasting packets can inform its presence to its
neighbors, the broadcasting packets such as Route REQuest
(RREQ) and Route ERRor (RERR) can play a role of Hello
packets.

The new proposed mechanism for reducing the routing
overhead is given in the Section III. The algorithm used for
reducing the routing overhead is given in the Section IV.

II. LITERATURE REVIEW

The routing overhead occurs because of the dissemination
of routing control packets such as RREQ packets can be large,
when the network topology changes frequently. The on-
demand routing protocols [13, 6] produce a large amount of
routing traffic by flooding the entire network with RREQ
packets in route discovery process. The issue of reducing the
7
th

routing overhead associated with route discovery and
maintenance in on demand routing protocols has attracted
the research work. Nelson et al [10] proposed a
methodology of dynamically adjusting the Hello timer
and the Timeout timer according to the conditions of the
network. In a high mobility network with frequent
topology changes the small values for the timers is used
to quickly detect the changes in the network. In a low
mobility network where the topology remains stable or
with the few changes, a large value of the timers is
effective to reduce the overhead. The reduction of
overhead is greatly achieved with the minimal cost of
slightly increasing the drop rate in data traffic. If the
packet loss increases by 1%, the overhead reduction
reaches 40%. Ould-Khaoua [9] proposed the two new
probabilistic route discovery method, called Adjusted
Probabilistic route discovery process (AP) and Enhance
Adjusted Probabilistic route discovery process (EAP)
which solves the broadcast storm problem in the existing
on-demand routing protocols. The probability for
forwarding is determined by the local density of the
sending node. In order to reduce the routing overhead
without degrading the network throughput in dense
network topology, the probability of packet forwarding
by the nodes located in sparse areas is set high while it is
set low at nodes located in dense areas. EAP-AODV
reduces overhead by 71% while APAODV reduces the
overhead by 55%.
Aminu [18] proposed a rebroadcast probability function
which takes in to account about the value of the packet
counter together with some key simulation parameters
like network topology size, transmission range and
number of nodes to determine the appropriate rebroadcast
probability for a given node. The rebroadcast probability
of the node is calculated based on the parameters.
Compared to the other mechanisms, simulation results
have revealed that counter Function achieved superior
saved rebroadcast about 20% better than its closest
competitor like counter-based scheme, in dense network
and end-to-end delay around 26% better than counter-
based scheme in dense network without sacrificing reach
ability in medium and dense networks. Kim [8] proposed
the probabilistic broadcasting method based on the
coverage area and the neighbor confirmation. This
method uses coverage area to set rebroadcast probability
and it uses the neighbor confirmation to make the
reachability. Haas [5] proposed the gossip approach in
which each node forwards a packet with probability. This
method can reduce to 35% overhead when compared to
flooding. Peng and Lu [12] proposed a neighbor
knowledge method named SBA- Scalable Broadcast
Algorithm. This method determines the rebroadcast of
packet, based on whether the rebroadcast can reach the
additional nodes. Abdulai [2] proposed Dynamic
Probabilistic Route Discovery (DPR) method based on
neighbor coverage method. In this method, each node
determines the probability for forwarding packets based
on the number of its neighbors and also the set of
neighbors which are covered by previous broadcast.
III. PROBABILISTIC REBROADCAST WITH
NEIGHBOR KNOWLEDGE (PRNK) MECHANISM

This paper proposes Probabilistic Rebroadcast with
Neighbor Knowledge (PRNK) mechanism which
combines both neighbor coverage and probabilistic
methods. In order to determine the neighbor coverage
knowledge, the rebroadcast delay is used to determine the
rebroadcast order, and the accurate coverage ratio is
obtained. To keep the network connectivity and to reduce
the redundant retransmissions, connectivity factor is used
to determine the number of neighbors to receive the
RREQ packet [17]. The coverage ratio and the
connectivity factor are combined finally. The upstream
coverage ratio of an RREQ packet received from the
previous node is used to calculate the rebroadcast delay
and the coverage ratio of the RREQ packet and the
connectivity factor is used to calculate the rebroadcast
probability. The rebroadcast probability is found, to
reduce the number of RREQ packet rebroadcasts and to
improve the routing performance.

3.1 REBROADCAST DELAY
This mechanism proposes a method
to determine the rebroadcast delay. The rebroadcast delay
is used to determine the forwarding order. The node
which has more common neighbors with the previous
node will have the lower delay. If the node rebroadcasts
the packet its common neighbors will know this
information. So, the rebroadcast delay used to spread the
information about the nodes that have disseminated the
packet to more neighbors. When a node ni receives
RREQ packet from its previous node ns, then node ns can
uses the neighbor list in the RREQ packet to determine
how many of its neighbors have not been covered by the
RREQ packet. If node ni has more neighbors that are not
covered by the RREQ packet from ns, then if node ni
rebroadcasts the RREQ packet, the RREQ packet will
reach more additional neighbor nodes. When node ns
sends an RREQ packet, all of its neighbors ni, where i =
1, 2 receive and process the RREQ packet. Assume the
node nk has the largest number of common neighbors
with node ns, and then node nk has the lowest delay. If the
node nk rebroadcasts the RREQ packet then there are
more nodes to receive the RREQ because node nk has the
largest number of common neighbors. The node nk
rebroadcasts the RREQ packet depends on the
rebroadcast

IV. ALGORITHM

The algorithm is given to reduce the routing
overhead in route discovery as follows:

RREQ s: RREQ packet received from node s.
Rs. id: the unique identifier (id) of RREQs.
N (u): Neighbor set of node u.
7
th

U (u, x): Uncovered neighbors set of node u for RREQ
whose id is x.
Timer (u, x): Timer of node u for RREQ packet whose id
is x.

ALGORITHM

if n
i
receives new RREQs from s then
Compute initial uncovered neighbors set U (n
i
, Rs.id)
for RREQs:

Compute the rebroadcast delay Td (n
i
):

Set a Timer (n
i
, Rs: id) according to
Td (n
i
) end if
while (n
i
receive a duplicate RREQ
j
from n
j
before
Timer (n
i
, Rs.id) expires) do
Adjust U (n
i
, Rs. id):

discard
(RREQ
j
)
end while
if Timer (n
i
, Rs. id) expires then
Compute the rebroadcast probability Pre (n
i
)

if Random (0, 1) Pre (n
i
) then

broadcast
(RREQs) else

Discard
(RREQs)
end if

K. PROTOCOL IMPLEMENTATION

To implement the proposed mechanism the
source code of AODV is enhanced in NS-2 [19, 20].
Figure 1 shows the data flow diagram for the
proposed model.

Existing AODV
model
Enhance AODV
model

Set up simulation with Set up simulation with
NS-2 NS-2

Comparison of the analysis

Figure 1 Data Flow Diagram

The proposed PRNK method needs Hello packets to get
the neighbor details and needs to carry the neighbor list
in RREQ packet. Therefore some methods are used to
reduce the overhead of Hello packets. In order to reduce
the overhead of Hello packets, the periodical Hello
mechanism is not used. Because when a node sends the
broadcasting packets, it can inform its existence to its
neighbors. The broadcasting packets like Route REQuest
(RREQ) and Route ERRor (RERR) play a role of Hello
packets. Each and every node monitors the neighbor table
and maintains the neighbor list in the received RREQ
packet in order to reduce the overhead of neighbor list in
RREQ packet. The neighbor table of any node n
i
has the
three cases to send or forward of RREQ packets:
1) The node n
i
sets the number of neighbors to a positive
integer if the neighbor table of node n
i
adds one new
neighbor

nj.
2) The node ni sets the number of neighbors to a negative
integer if the neighbor table of node n
i
deletes any of its
neighbors.

3) The node n
i
does not need to list its neighbors if the
neighbor table of node n
i
does not vary and set the
number of neighbors to 0.
The nodes when receives the RREQ packet from node n
i

can take their actions according to the value of num
neighbors in the received RREQ packet:-

i. If the number of neighbors is a positive integer
then node substitutes its neighbor cache of node
n
i
according to the neighbor list in the received
RREQ packet.

7
th

ii. If the number of neighbors is a negative integer
then node updates its neighbor cache of node n
i

and deletes the deleted neighbors in the
received RREQ packet.
iii. If the number of neighbor is 0 then the node
does nothing.

Due of the case ii and case iii this method reduces the
overhead of neighbor list listed in the RREQ packet.

The performance evaluation of the proposed mechanism
is measured using the following performance metric.

Normalized routing overhead is defined as the ratio of
the total packet size of control packets including RREQ,
RREP, RERR and Hello packet to the total packet size
of data packets delivered to the destinations. The size of
RREQ packets is used instead of the number of RREQ
packets, since the protocols include a neighbor list in the
RREQ packet so that its size is bigger than the original
AODV.

5.1 SIMULATION PARAMETERS

The simulation is conducted using Network
Simulator NS2. The simulation parameters are
tabulated in Table.1

Table1. Simulation Parameters

SIMULATION VALUES
PARAMETER

Simulator NS-2 (v2.34)

Topology Size 800m x 800m

Number of nodes 80,90,100,150

Bandwidth 2Mbps

Transmission Range 250m

Interface Queue Length 50

The MAC layer protocol uses the Distributed
Coordination Function (DCF) of IEEE 802.11 protocol.
The radio channel model is Lucents WaveLAN. The
traffic type used is Constant Bit Rate (CBR) and the
source- destination connection is chosen randomly.

The initial simulation for AODV protocol shows the
following result for routing overhead in Figure 1.
Figure 2 Routing Overhead

The AODV protocol is modified to implement the PRNK
mechanism and the results are compared between the
AODV protocol and the proposed mechanism.
VI. CONCLUSION

Broadcasting is the basic mechanism used in
reactive routing protocols. The important issue is to
reduce the number of rebroadcast packets. This paper
proposes a new Rebroadcast Technique for Reducing
Traffic Type CBR (Constant Bit Rate)

Packet Size 512 bytes

Pause Time 200s

7
th

Routing Overhead in Mobile Ad Hoc Networks
(MANETs). The neighbor coverage knowledge
comprises of coverage ratio and the connectivity factor.
This technique dynamically calculates the rebroadcast
delay, which finds the forwarding order and also
effectively find the neighbor coverage knowledge. The
result of the simulation shows that the new mechanism
produces less rebroadcast than the existing protocol. Due
to less redundant rebroadcast, the proposed mechanism
reduces the network collision and contention so that it
may also increases the packet delivery ratio and also
decreases the average end-to-end delay. The simulation
result shows that the new mechanism has better
performance when the network is in high density. The
future work is focused on calculating the result for
performance metrics like MAC collision rate.

REFERENCES
[1] H. AlAamri, M. Abolhasan, and T. Wysocki, On
Optimising
Route Discovery in Absence of Previous Route
Information in
MANETs, Proc. IEEE Vehicular Technology Conf.
(VTC), pp.-5, 2009.
[2] J.D. Abdulai, M. Ould-Khaoua, L.M. Mackenzie,
and A.
Mohammed, Neighbour Coverage: A Dynamic
Probabilistic Route Discovery for Mobile Ad Hoc
Networks, Proc. Intl Symp.
Performance Evaluation of Computer and Telecomm.
Systems
(SPECTS 08), pp. 165-172, 2008.
[3] J.D. Abdulai, M. Ould-Khaoua, and L.M. Mackenzie,
Improving Probabilistic Route Discovery in Mobile Ad
Hoc Networks, Proc.
IEEE Conf. Local Computer Networks, pp. 739-746,
2007.
[4] J. Chen, Y.Z. Lee, H. Zhou, M. Gerla, and Y. Shu,
Robust Ad Hoc Routing for Lossy Wireless
Environment, Proc. IEEE Conf. Military Comm.
(MILCOM 06), pp. 1-7, 2006.
[5] Z. Haas, J.Y. Halpern, and L. Li, Gossip-Based Ad
Hoc Routing, Proc. IEEE INFOCOM, vol. 21, pp. 1707-
1716, 2002.
[6] D. Johnson, Y. Hu, and D. Maltz, The Dynamic
Source Routing Protocol for Mobile Ad Hoc Networks
(DSR) for IPv4, IETF RFC 4728, vol. 15, 2007.
[7] A. Keshavarz-Haddady, V. Ribeirox, and R. Riedi,
DRB and
DCCB: Efficient and Robust Dynamic Broadcast for Ad
Hoc and Sensor Networks, Proc. IEEE Comm. Soc.
Conf. Sensor, Mesh, and Ad Hoc Comm. and Networks
(SECON 07), pp. 253-262, 2007.
[8] J. Kim, Q. Zhang, and D.P. Agrawal, Probabilistic
Broadcasting
Based on Coverage Area and Neighbor Confirmation in
Mobile Ad Hoc Networks, Proc. IEEE GlobeCom,
2004.
[9] A. Mohammed, M. Ould-Khaoua, L.M. Mackenzie,
C. Perkins, and J.D. Abdulai, Probabilistic Counter-
Based Route Discovery for Mobile Ad Hoc Networks,
Proc. Intl Conf. Wireless Comm. and
Mobile Computing: Connecting the World Wirelessly
(IWCMC 09), pp. 1335-1339, 2009.
[10] Nelson, Shoji Yutaka, Takahashi.2010. Dynamic
Hello/Timeout timer adjustment in routing protocols for
reducing overhead in MANETs
[11] S.Y. Ni, Y.C. Tseng, Y.S. Chen, and J.P. Sheu, The
Broadcast Storm Problem in a Mobile Ad Hoc Network,
Proc. ACM/IEEE

MobiCom, pp. 151-162, 1999.
[12] W. Peng and X. Lu, On the Reduction of
Broadcast Redundancy in Mobile Ad Hoc Networks,
Proc. ACM MobiHoc, pp. 129-130, 2000.
[13] C. Perkins, E. Belding-Royer, and S. Das, Ad Hoc
On-Demand Distance Vector (AODV) Routing, 2003.
[14] F. Stann, J. Heidemann, R. Shroff, and M.Z.
Murtaza, RBP: Robust Broadcast Propagation in
Wireless Networks, Proc. Intl Conf. Embedded
Networked Sensor Systems (SenSys 06), pp. 85-98,
2006.

B. Williams and T. Camp, Comparison of
Broadcasting Techniques for Mobile Ad Hoc Networks,
Proc. ACM MobiHoc, pp. 194- 205, 2002.
X. Wu, H.R. Sadjadpour, and J.J. Garcia-Luna-
Aceves,

Routing Overhead as a Function of Node Mobility:
Modeling Framework and Implications on Proactive
Routing, Proc. IEEE Intl Conf. Mobile Ad Hoc and
Sensor Systems (MASS 07), pp. 1- 9, 2007.
F. Xue and P.R. Kumar, The Number of Neighbors
Needed for Connectivity of Wireless Networks,
Wireless Networks, vol. 10, no. 2, pp. 169-181, 2004.
An Improved Rebroadcast Probability Function for
an Efficient Counter-Based Broadcast Scheme in
MANETs

[19] The Network Simulator (NS2) website,
http://www.isi.edu/nsnam/ns/ns-build.html
[20] The Network Simulator manual,
www.isi.edu/nsnam/ns/ns-documentation.html
7
th

REDUCTION OF POWER DISSIPATION OVER THE MANET
THROUGH HELLO MESSAGING SGHEME WITH
ONDEMAND ROUTING PROTOCOLS

A.Anitha. K.Naveen Durai
Student, Department of Computer Science Assistant professor, Department of Computer Science
Sri Eshwar College Of Engineering Sri Eshwar College Of Engineering
Coimbatore, India Coimbatore, India
aspireanitha@gmail.com srmnaveen18@gmail.com

Abstract -
A mobile adhoc network is a
collection of many mobile nodes that communicate
with one another without any fixed networking
infrastructure. In this network the nodes are
mobile, the power management and energy
conversation become very critical in mobile adhoc
network. In mobile Adhoc network the
discovering the neighbor node is important
consideration for data transferring from source
node to designation node. The existent method of
hello messaging scheme is rapidly consume the
power in batteries while the mobile are not in use.
The paper proposes a scheme for suppresses the
unnecessary hello messages by fixing the constant
time interval for probability of finding the broken
links as event instead of constant time interval of
hello messages. This will lead to increase the life
time of battery in mobile nodes,

Keywords: Manet, Hello Messaging Scheme, Hello
interval. Event interval

I.INTRODUCTION

AD-HOC mobile phone networks have gained
attention as smart phones and various applications
such as ftp server [1] are deployed widely. In a
mobile ad-hoc network (MANET) using smart
phones, energy efficiency is a major concern
(In.Fig.1). Pathak et al. show no-sleep energy bugs
can entirely drain batteries while phones are not in
use [2].

Discovery of neighbor nodes can also be a
hidden energy drain in ad-hoc mode. In MANETs,
any node in a route can move away or be turned off,
which negatively affects route maintenance and
throughput may cause delays in data dissemination,
and so on. It is vitally important for a node in a
MANET to discover live neighbor nodes through
Hello messaging or a link layer feedback mechanism.
For neighbor discovery, periodically exchanging.

Fig .1 Ad-hoc Networks

1.1 HELLO MESSAGES

Network connectivity may be determined
through the reception of broadcast control messages.
Any broadcast control message also serves as a hello
message, indicating the presence of a neighbor. When
a node receives a hello message from its neighbor, it
creates or refreshes the routing table entry to the
neighbor .To maintains connectivity, if a node has
not sent any broadcast control message within a
specified interval, a hello message is locally
broadcast. This results in at least one hello message
transmission during every time period. Failure to
7
th

receive any hello message from a neighbor for
several time intervals indicates that neighbor is no
longer within transmission range, and connectivity
has been lost. Two variables control the
determination of connectivity using hello messages:
HELLO INTERVAL and ALLOWED HELLO
LOSS. HELLO INTERVAL specifies the maximum
time interval between the transmissions of hello
messages. ALLOWED HELLO LOSS specifies the
maximum number of periods of HELLO INTERVAL
to wait without receiving a hello message before
detecting a loss of connectivity to a neighbor. The
recommended value for HELLO INTERVAL is one
second and for ALLOWED HELLO LOSS is two
[11]. In other words, if a hello message is not
received from a neighbor within two seconds of the
last message, a loss of connectivity to that neighbor is
determined.

1.2 ROUTE DISCOVERY

Fig .2 Hello-Messages

When a source needs to send packets to a
destination (In Fig.2), it first must determine a path
for communication. The source node begins route
discovery by broadcasting a route request (RREQ)
message containing the IP address of the destination.
When an intermediate node receives the RREQ, it
records the reverse route toward the source and
checks whether it has a route to the destination.

If a route to the destination is not known, the
intermediate node rebroadcasts the RREQ.RREQ
propagation is illustrated in Figure 1(b). When the
destination, or an intermediate node with recent
information about a route to the destination, receives
the RREQ, a route reply (RREP) is generated. The
RREP is unicast back to the source using the reverse
route created by the RREQ. For example, in Figure
1(c) two nodes have recent information about the
destination because hello messages are being used.
These two nodes unicast a RREP to the source. As
the RREP propagates toward the source, a forward
route to the destination is created at each
intermediate hop. When a RREP reaches the source,
the source records the route to the destination and
begins sending data packets to the destination along
the discovered path, as illustrated in Figure 1(d). If
more than one RREP is received by the source, the
route with the lowest hop count to the destination is
selected.

1.3 ROUTE MAINTENANCE

When a link breaks along an active path, the
node upstream of the break detects the break
message. The RERR message lists all destinations
that are now unreachable, due to the link break. The
node then sends the RERR message toward the
source. Each intermediate hop deletes any broken
routes and forwards the RERR packet toward the
source. When the source receives the RERR packet it
determines whether it still needs the route to the
destination. If so, the source creates a RREQ and
begins the route discovery process again.

The Dynamic MANET On-demand (DYMO)
routing protocol enables reactive, multihop routing
between participating nodes that wish to
communicate. The basic operations of the DYMO
protocol are route discovery and management.
During route discovery the originating node initiates
dissemination of a Route Request (RREQ)
throughout the network to find the target node.
During this dissemination process, each intermediate
node records a route to the originating node. When
the target node receives the RREQ, it responds with a
Route Reply (RREP) unicast toward the originating
node. Each node that receives the RREP records a
route to the target node, and then the RREP is unicast
toward the originating node. When the originating
node receives the RREP, routes have then been
established between the originating node and the
target node in both directions.

In order to react to changes in the network
topology nodes maintain their routes and monitor
their links. When a packet is received for a route that
is no longer available the source of the packet is
notified. A Route Error (RERR) is sent to the packet
source to indicate the current route is broken. Once
the source receives the RERR, it re-initiates route
discovery if it still has packets to deliver. In order to
7
th

enable extension of the base specification, DYMO
defines a generic element structure and handling of
future extensions. By defining a fixed structure and
default handling, future extensions are handled in a
predetermined fashion. DYMO uses sequence
numbers as they have been proven to ensure loop
freedom

IP Destination Address (IP Destination Address)
The destination of a packet, indicated by examining
the IP header.IP Source Address (IPSourceAddress)
The source of a packet, indicated by examining the IP
header. DYMO cast Packet transmission to all
DYMO routers. DYMOcast packets should be sent
with an IP Destination Address of IPv4 TBD (IPv6
TBD), the DYMO cast Address .Routing Element
(RE)A DYMO message element that is used to
distribute routing information. Route Invalidation
Disabling the use of route, causing it to be
unavailable for forwarding data. Route Reply (RREP)
upon receiving a RREQ, the target node generates a
Route Reply (RREP). A RREP is a RE with a unicast
IP Destination Address, indicating that this RE is to
be unicast hop-by-hop toward the Target Address.
Route Request (RREQ) A node generates a Route
Request (RREQ) to discover a valid route to a
particular destination (Target Address). A RREQ is
simply a RE with the DYMO cast Address in the IP
Destination Address field of the IP packet. Also, the
A-bit is set to one (A=1) to indicate that the Target
Node must respond with a RREP. Valid Route A
known route where the Route. Valid Timeout is
greater than the current time.

Hello messages are preferred over link layer
feedback because the former does not restrict usage
and implementation to a specific link layer
technology such as ACK packets [3].

Many Hello messaging schemes focus on
figuring out dynamic network topology [4] or
discovering live neighbors with an energy saving
scheme [5], which requires all network nodes to
continuously exchange Hello messages or beacons
while they are awake. In such traditional Hello
messaging schemes no start/end condition is
described [6]. This can cause unnecessary bandwidth
usage and hidden energy consumption if an on-
demand MANET routing protocol (e.g., Ad hoc On-
Demand Distance Vector (AODV) [7], or Dynamic
MANET On-demand (DYMO) [8]) is used, where a
new path is discovered through Route Request
(RREQ) and Route Response (RREP) packet
exchanges.

Giruka and Singhal proposed two approaches
for suppressing Hello messages when they are not
required [9]:an on-demand mechanism (reactive
Hello protocol), and a monitoring activity mechanism
(event-based Hello protocol).Source-initiated on-
demand protocols create routes only when these
routes are needed. The need is initiated by the source,
as the name suggests. When a node requires a route
to a destination, it initiates a route discovery process
within the network. The reactive Hello protocol
enables Hello messaging only when it is demanded
using a Hello request-reply mechanism, but increases
delay due to additional packet exchange before
communication. The event-based Hello protocol
enables only active nodes (i.e., those either sending
or receiving data packets) to broadcast Hello packets
based on a threshold called an activity timer.
However, a threshold that is set too high rarely
reduces the Hello messaging overhead, whereas a
low threshold results in local connectivity
information loss.

Thus, there is an outstanding need to effectively
suppress unnecessary Hello messaging while
minimizing the risk of losing local connectivity
information. In this paper, we propose an adaptive
Hello messaging scheme for neighbor discovery by
effectively suppressing un-necessary Hello messages.
The proposed scheme dynamically adjusts Hello
intervals, and does not increase the risk that a sender
will transmit a packet through a broken link that has
not been detected by Hello messaging; we call this
the probability of failure of detection of an
unavailable link (P
F D
). To estimate this risk, we
exploit an average event interval, that is, an average
time gap between two consecutive events (i.e.,
sending or receiving a data packet) on a node. By
monitoring the event intervals, we can estimate how
actively a node is involved in sending or forwarding.
If a node is not involved in any communication for a
given period, it does not need to maintain the status
of the link; Hello packets broadcasted during this
period are unnecessary. If a constant Hello interval is
used, the risk of attempting to transmit a packet
through a broken link decreases as the event interval
increases. Instead of using a constant Hello interval,
our proposed scheme uses a constant risk level.

As the event interval increases, the Hello
interval can also increase without increasing risk. If
the event interval is extremely large, the Hello
messaging interval is also correspondingly large; that
is Hello messaging is practically suppressed. When a
node receives or sends a packet, the Hello messaging
interval is reset to a default value so that up-to-date
information is kept in a neighbor table for active
communication. Simulation results show that our
proposed scheme suppresses unnecessary Hello
messaging and reduces the energy consumption up to
7
th

54% without any additional delay.

II. DESIGN OF AN ADAPTIVE HELLO
MESSAGING INTERVAL SCHEME

The basis of using hello messages to determine
connectivity stems from the assumption that
reception of a hello message indicates a viable
communication channel with the source of the hello.
This mechanism works well on wired networks,
which experience few packet losses and connectivity
changes. However, when used in wireless ad hoc
networks the effectiveness decreases due to many
factors. Some of the factors that have significant
effect are: hello loss settings, hello packet size and
802.11b packet handling, A variety of approaches for
improving the accuracy of hello messages as an
indicator of local connectivity are examined.

The ideal behavior of a network regarding local
link connectivity is that if a node moves away, its
neighbor nodes recognize a link failure before a
packet to send arrives. Figure 3 depicts the time
analysis. T
d
represents the time for link

Fig. 3. The time parameters of the link connectivity.

failure detection based on periodic Hello packet
messaging. The average T
d
is given as
T
d
= (ALLOWED HELLO LOSS0.5)*HELLO
INTERVAL

Fig.3 represents a conventional scheme updates
link availability information as soon as it recognizes
a link failure with one of its neighboring nodes.
However, link failure information does not actually
need to be updated until its neighbor is involved in
communication such as forwarding a packet. T
w

represents the time interval between when a link
failure is recognized and when the link is actually
needed. During T
w
, multiple Hello packets are
superfluous. Moreover, if a node has moved into an
area where no active nodes are in its neighborhood
and keeps broadcasting Hello packets, energy is
consumed unnecessarily. To avoid this problem,
Hello packets should be suppressed. The proper
solution, however, depends on determining the
correct Hello interval.

Let us consider a sender and its neighboring
node. An event on the neighboring node occurs when
the sender forwards a packet to the neighboring node.
If the neighboring node moves out of the senders
transmission range, there are two possible courses of
action: (1) the sender is requested to forward; or (2)
the sender is not requested to forward.

Here, only case (1) can incur a link error. In
case (2), the link availability does not need to be
updated. To prevent a link error in the former case,
the sender must know the availability of its link to the
next hop node prior to forwarding a packet. In other
words, the link connectivity information from the
next hop node obtained by the last received Hello
packet should be valid or refreshed before a packet is
forwarded. Setting the HELLO INTERVAL without
considering the event interval of a node can increase
the risk of sending a packet through an unavailable
link. The goal of this paper is to find the expiration
period corresponding to a given event interval so that
unnecessary Hello messages are suppressed. Since a
Hello packet is un-necessary in case (2), the
probability of a node forwarding a packet in a given
time period needs to be estimated in order to avoid
the outcome in case (1). We investigated the
distribution of packet reception event intervals on
each node when the simulation parameters in Table I
with 50 nodes are used. File sharing over TCP and
VoIP over UDP are practical applications in phone-
to-phone network [10]. A ftp traffic generator is used
for TCP while a pareto traffic generator is for UDP
because UDP traffic on realistic wireless networks
can be modeled by pareto to represent sequences of
almost stationary periods [11]. We observe most
event intervals are less than a default HELLO
INTERVAL (i.e., 1 sec in [7]), however, we
investigate the event intervals, x, that are larger than
1 sec because the Hello interval is not modified if the
intervals less than the default HELLO INTERVAL.
The cumulative distributed function (CDF) in Figure
2 shows that all traffics are bounded by the
exponential distribution where x > 1. The CDF of x is
as follows:

F (x, ) = 1 e
x/
(2)

The exponential distribution implies that

7
th

portion of intervals are less than a given interval x.
This means that if we use an exponential distribution
to estimate the probability that an interval is larger
than a given interval, the estimation will give an
upper bound. So the probability that an event occurs
after a given interval is negligible. We use this
analysis to adjust Hello interval conservatively. We
can interpret x as a link refresh period (T
d
) and F (x,
) as the probability that an event occurs before the
link is refreshed, i.e., F (x, ) = P
F D
. Since a
conventional Hello messaging scheme uses a
constant value for, P
F D
varies depending on . This
causes even inactive nodes to broadcast Hello packets
periodically. We fix (=P
F D
) and make a variable so
that the Hello interval is adaptive to the average event
interval of a node, . We can rewrite the CDF of the
exponential distribution using the P
F D
as follows:

1-
e
-x/5
= P
FD

Note that x increases linearly as increases.
The neigh-boring node in the aforementioned two
cases can use for calculating the appropriate Hello
interval to maximize Td and minimize T
w
in Figure 1.
Once T
d
is determined, the Hello messaging interval
of the neighboring node can be calculated with
equation 1. With real traffic, the risk of sending a
packet over a broken link will be less than P
F D
in
equation 3 because P
F D
is the upper bound obtained
by an exponential distribution.

III. EVALUATION
This section shows the effect of the proposed
scheme on energy use, throughput, and Hello packet
ratio for a data packet with Ns-2 v.2.35 within
various simulation parameters including node
density, number of flows, mobility speed, and P
F D

.We modified AODV and DYMO with the proposed
scheme, which we called AODV with adaptive Hello
(AODV-AH) and DYMO with adaptive Hello
(DYMO-AH), respectively. The average interval
parameter in equation 3 is obtained by monitoring
the time interval since the last packet was received.
The default P
F D
is set to 20%. We used 100 different
network topologies. Each point in each figure
represents the average of 30 simulation runs. Unless
explicitly stated otherwise, the simulation parameters
shown in Table I are used. For pareto flows, packet
size is 210 bytes, burst and idle time are 500ms, and
the shape parameter is 1.5.

3.1 ROUTE TABLE ENTRY

The route table entry is a conceptual data
structure. Implementations may use any internal
representation that conforms to the semantics of a
route as specified in this document.

- Route. DestAddress
- Route. DeleteTimeout
- Route.HopCnt
- Route.IsGateway
- Route.NextHopAddress
- Route.NextHopInterface
- Route.Prefix
- Route.SeqNum
- Route.ValidTimeout

These fields are defined as follows:

Route Node Address (Route.
DestAddress).The IP address of the node associated
with the routing table entry. Route Delete Timeout
(Route. DeleteTimeout) If the time current is after
Route. DeleteTimeout the corresponding routing
table entry MUST be deleted. Route Hop Count
(Route.HopCnt) The number of intermediate node
hops before reaching the Route.DestAddress.Route Is
Gateway (Route.IsGateway)1-bit selector indicating
whether the Route. DestAddress is a gateway. Route
Next Hop Address (Route.NextHopAddress)The IP
address of the next node on the path toward the
Route.DestAddress.Route Next Hop Interface
(Route.NextHopInterface)The interface used to send
packets toward the Route.DestAddress.Route Prefix
(Route.Prefix)6-bit field that specifies the size of the

subnet reachable through the Route.DestAddress,The
definition of the Prefix field is different for gateways;
entries with Route. Is Gateway set to one (1).Route
Sequence Number(Route.SeqNum) The sequence
number of the
Route.DestAddress.Route.ValidTimeout The time at
which a route table entry is scheduled to be
invalidated. The routing table entry is no longer
considered valid if the current time is after
Route.ValidTimeout.
3.2 DYMO ROUTING TABLE OPERATIONS.
While processing a RE, as node checks its routing
table for an entry to the RBNodeAddress using
longest-prefix matching. In the event that no
matching entry is found, an entry is created.
If a matching entry is found, the routing information
about RBNodeAddress contained in this RBlock is
considered stale if:
7
th

- The result of subtracting the Route.SeqNum
from RBNodeSeqNum is less than zero (0)
using signed 32-bit arithmetic, OR
- The result of subtracting the Route.SeqNum
from RBNodeSeqNum is equal to zero (0)
using signed 32-bit arithmetic AND the
RBHopCnt is greater than Route.HopCnt.
If there exists a route AND the result of subtracting
the Route.SeqNum from RBNodeSeqNum is equal to
zero (0) using signed 32-bit arithmetic AND the
RBHopCnt is equal to the Route.HopCnt in this
RBlock the information is not stale, but the routing
information SHOULD be disregarded and no routing
update should occur.
If the information in this RBlock is stale or
disregarded and this RBlock is the first node in the
RR this DYMO packet MUST be dropped. For other
RBlocks containing stale or disregarded routing
information, the RBlock is simply removed from this
RE and the RELen adjusted. Removing stale and
disregarded RBlocks ensures that unused information
is not propagated further.
Fig.4 examines the distribution of the event interval
over ftp and pareto. In generally the periodic time is
always less than 1 sec ,so comparison between the ftp
and pareto in distribution of event interval are show
in the graph
.
Fig. 4. Distribution of event intervals (> 1sec).

TABLE I
SIMULATION PARAMETERS

Parameter Value Parameter
Valu
e
Propagatio
n
Rayleigh
fad- Mobility Max 5 m/s,
model ing speed
0
sec
paus
e
time
Transmissi
on 16dBm Mobility
Random
way
power model point
Topology
size
670 x 670
m Flows ftp, Pareto
Data rate 11Mbps
Number
of 10
senders
Number
of 20
PHY/MA
C
802.
11 (no
nodes
RTS
)

Fig. 5. DYMO-AH extends the battery life time.

Fig.5 compares the energy consumption
between DYMO and DYMO-AH by showing the
average of remaining energy of all nodes over time
when there are variable ftp flows. Each node initially
has 150 joules energy quickly due to periodic
reception and transmission of Hello packets. When
there is only one flow, this hidden energy
consumption causes unnecessary battery drain in
most nodes that are not involved in any
communication. As the proposed scheme reduces the
number of unnecessary Hello packets, energy is
saved due to less transmission and less receptions.
We investigated the throughput of ftp over tcp and
pareto flows and energy consumption when the
number of flows varies. The number of nodes is set to
50, and a number of flows (5, 10, 15, 20) are
randomly selected. No throughput drop is observed
which shows that the proposed scheme does not
reduce the detectability of a broken link; that is, only
7
th

unnecessary Hello packets are suppressed. Figure 4
depicts the consumed energy per received packet
over various numbers of flows at t = 60 minutes. The
effect of energy saving is high when the number of
flows is less than 5 (which is typical in Wife [12]).
As the number of flows increases, the effect of the
energy saving decreases because more nodes will
participate in forwarding.

3.4 CONFIGURATION PARAMETERS

Here are some default parameter values for
DYMO: Parameter Name Suggested Value

NET_DIAMETER 10
RATE_LIMIT 10
ROUTE_TIMEOUT 3000 milliseconds
ROUTE_DELETE_TIMEOUT
5*ROUTE_TIMEOUT
RREQ_WAIT_TIME 1000 milliseconds
RREQ_TRIES 3

For large networks or networks with frequent
topology changes the default DYMO parameters
should be adjusted using either experimentally
determined values or dynamic adaptation. For
example, in networks with infrequent topology
changes ROUTE_TIMEOUT may be set to a much
larger value It is assumed that all nodes in the
network share the same parameter settings. Different
parameter values for ROUTE_TIMEOUT or
ROUTE_DELETE_TIMEOUT in addition to
arbitrary packet delays may result in frequent route
breaks or routing loops. When a node processes a
RERR after generic element pre-processing, it
SHOULD set the Route.ValidTimeout to the current
time for each route to a UNodeAddress that meets all
of the following conditions:
1. The Route.NextHopAddress is the same as
the RERR IPSourceAddress.
2. The Route.NextHopInterface is the same as
the interface on which the RERR was
received.
3. The UNodeSeqNum is zero (0) OR the
result of subtracting Route.SeqNum from
UNodeSeqNum is less than or equal to zero
using signed 32-bit arithmetic.
Each UNodeAddress that did not result in a change to
Route.ValidTimeout SHOULD be removed from the
RERR.

Fig.6 Energy consumption for variable flows.

Fig.7 Hello packet overhead.

Fig.8 Throughput for various max speed and P
FD
.

The proposed scheme decreases the number of
Hello packets by as much as half. (Fig.7)The effect of
the proposed scheme increases as the number of
nodes increases. This is because the number of Hello
packet broadcasters and the number of received Hello
packets by a node also increase as the number of
nodes does. Due to the shortest hop policy in routing
protocols, the number of intermediate nodes rarely
changes regardless of node density and (fig.6)
compares the number of various flows with
consumed energy this will clear the idea about how
the energy consumed for various flows. Thus, only a
7
th

small portion of neighboring nodes are involved in
communication, and the other nodes will
correspondingly increase their Hello intervals. To
enable this scheme, each Hello packet from a next
hop node must include 1 byte of Hello interval
information in order to inform its senders, which is
negligible compared with the reduced amount of the
Hello packets.

Before a route can be used for forwarding a
packet, it MUST be checked to make sure that the
route is still valid. If the Route.ValidTimeout is
earlier than the current time, the packet cannot be
forwarded, and a RERR message MUST be
generated. In this case, the Route. DeleteTimeout is
set to Route.ValidTimeout +
ROUTE_DELETE_TIMEOUT.

If the current time is after Route. DeleteTimeout,
then the route SHOULD be deleted, though a route
MAY be deleted at any time.
Nodes MUST monitor links on active routes. This
may be accomplished by one or several mechanisms.
Including:
- Link layer feedback
- Hello messages
- Neighbor discovery
- Route timeout
Upon detecting a link break the detecting node
MUST set the Route.ValidTimeout to the current
time for all routes active routes utilizing the broken
link.
A RERR MUST be issued if a data packet is received
and it cannot be delivered to the next hop. RERR. A
RERR SHOULD be issued after detecting a broken
link of an active route to quickly notify nodes that a
link break occurred and a route or routes are no
longer available.

To avoid route timeouts for active routes, a
node MUST update the Route.ValidTimeout to the
IPSourceAddress to be the current time +
ROUTE_TIMEOUT upon receiving a data packet.
To avoid route timeouts for active routes, a node
SHOULD update the Route.ValidTimeout to the
IPDestinationAddress to be the current time +
ROUTE_TIMEOUT upon successfully transmitting a
packet to the next hop. A (DYMO-AH) packet may
consist of multiple DYMO elements. Each element is
processed individually and in sequence, from first to
last. (Fig.8) shows the impact of P
F D
on the
throughput and Hello ratio when the max speed
varies. A high P
F D
uses a longer Hello interval than a
low P
F D
. Surprisingly, a high P
F D
does not make
significant difference in throughput from

Fig. 9. CDF of active nodes.

a low P
F D
. This is because the number of affected
links is small; a link will be affected by a high P
F D

only when there is a forwarding event before the link
is refreshed and the neighboring node moves away.
However, since most of event intervals are small,
only a few mobile nodes (less than 1%) that happen
to be involved in forwarding will temporarily
experience a very large Hello interval due to a high
P
F D
. Moreover, although the first several packets
may be lost in circumstances where P
F D
is high, the
Hello interval is reset to the default value at the first
packets arrival, and the path recovery mechanism
will invocated immediately. Successive packets will
be forwarded through a valid link. That is why a high
P
F D
does not significantly decrease throughput
significantly. The same explanation accounts for the
impact of the high loss of Hello packets due to poor
link quality. To show the impact of the link quality
on the proposed scheme, we vary transmission
power. The overall throughput decreases due to poor
link quality, however, the proposed scheme still
shows no throughput drop compared with the existing
scheme owing to the quick reset to a default value.

Fig.9 shows the number of nodes that will
increase their Hello intervals. The cumulative
probability of active nodes that send or receive
packets during recent x seconds (x=3, 5, and 10 sec)
are shown. The number of flows is represented on the
x-axis. When there are 10 ftp flows and x=3sec, 85%
of nodes do not send or receive in the most recent 3
seconds i.e., there is no need refresh within 3
seconds. This indicates that a large portion of nodes
use a longer Hello interval than the current default.
IV. CONCLUSION

In this paper, we proposed a new scheme for
reducing the power dissipation of power over the
7
th

Manet by the combined approaches of traditional
hello messaging scheme with reactive protocols like
AODV and DYMO, an adaptive Hello interval to
reduce battery drain through practical suppression of
unnecessary Hello messaging. Based on the event
interval of a node, the Hello interval can be enlarged
without reduced detectability of a broken link, which
decreases network overhead and hidden energy
consumption.

REFERENCES

1. Mochasoft, Mocha FTP server. Available:
http://www.mochasoft.dk/ freeware/ftpd.htm.
2. A. Pathak, A. Jindal, Y. C. Hu, and S. P.
Midkiff, What is keeping my phone awake?
Characterizing and detecting no-sleep energy
bugs in smartphone apps, in Proc. 2012
International Conference on Mobile Systems,
Applications, and Services, pp. 267280.
3. C. Gomez, M. Catalan, X. Mantecon, J.
Paradells, and A. Calveras, Evaluating
performance of real ad-hoc networks using
AODV with hello message mechanism for
maintaining local connectivity, in Proc. 2005
IEEE International Symposium on Personal,
Indoor and Mobile Radio Communications, vol.
2, pp. 13271331.

4. R. Oliveira, M. Luis, L. Bernardo, R. Dinis, and
P. Pinto, The impact of nodes mobility on link-
detection based on routing hello messages, in
Proc. 2010 IEEE Wireless Communications and
Networking Conference, pp16.
5. C.-M. Chao, J.-P. Sheu, and I.-C. Chou, An
adaptive quorum-based energy conserving
protocol for IEEE 802.11 ad hoc networks,
IEEE Trans. Mobile Computing, vol. 5, no. 5,
pp. 560570, May 2006.
6. T. Clausen, C. Dearlove, and J. Dean, Mobile
ad hoc network (MANET) neighborhood
discovery protocol (NHDP), 2010.
7. E. Belding-Royer and S. D. C. Perkins, Ad hoc
on-demand distance vector (AODV) routing,
July 2003.

8. I. D. Chakeres and C. E. Perkins, Dynamic
MANET on-demand (AODVV2) routing, Feb.
2008.

9. V. C. Giruka and M. Singhal, Hello protocols
for ad-hoc networks: overhead and accuracy
tradeoffs, in Proc. Sixth IEEE International
Symposium on a World of Wireless Mobile and
Multimedia Networks,
a. 354361.

10. G. Iannello, F. Palmieri, A. Pescape, and P. S.
Rossi, End-to-end packet-channel Bayesian
model applied to heterogeneous wireless
networks, in Proc. 2005 IEEE Global
Telecommunications Conference, pp. 484489.
11. J. Hui and M. Devetsikiotis, The use of
metamodeling for VoIP over WiFi capacity
evaluation, IEEE Trans. Wireless Commun.,
vol. 7, no. 1, pp15, Jan. 2008.

7
th

PREVENTING COLLABORATIVE ATTACKS USING COOPERATIVE
IMMUNE SYSTEM AND RBF NEURAL NETWORK IN MANET

M.Revathi
Assistant Professor,
Paavai Engineering College, Namakkal
revathi.jul27@gmail.com

K.Prakash
Assistant Professor,
Paavai Engineering College, Namakkal
prakashgreen@gmail.com

Abstract - The majority of the present security
systems do not provide enough level of protection
against collaborative attacks such as black hole
attacks and wormhole attacks, in the mobile ad hoc
networks. The main reason for their failure is the use
of point solutions to protect hosts and reactive
approach against intrusions. Collaborative attacks in
the mobile ad hoc networks can cause more damages
than individual attacks. Inspired by the human
immune system, a tri-tier (i.e., Native, Parallel, and
Adaptive) cooperative immune model is introduced
and collaborative attacks in the mobile ad hoc
networks. In Adaptive immune tier, A Multi Agent
system has been adopted in the immune learning to
perform detection in an efficient manner using RBF
neural network. Based on the output, a counter attack
is dispatched to surround the invaders and kill it. Our
solution overcomes the limitations of traditional
security solutions and providing protection against the
collaborative attacks
Keywords Immune system, Agent, RBF

Introduction
Security is a challenge issues in the mobile adhoc
networks. But, the protocols of these networks make
the vulnerabilities available to attackers. However, the
mobile adhoc network is not secure against
collaborative attacks because the security approaches
are suitable for only individual attacks. Collaborative
attacks are launched by some malicious adversaries to
accomplish disruption, deception, usurpation, or
disclosure against the targeted networks [1]. To deal
with the collaborative attacks, some cooperative
approaches are designed and used for matching the
features of multiple attacks in collaborative ways.
Unfortunately, these approaches are often ineffective
to unknown attacks [3]. In fact, human immune
network is an advanced natural cooperative defending
system against collaborative attacks from viruses,
bacteria, and cancer. Bacteria with the viruses and
cancer can cause the overload and damages of the
immune system.
In recent years, biological inspired techniques have
been used successfully to solve many complex and
computational problems. In this paper, how artificial
immune system coupled with the agent based
technology can realistically improve the security of
large and complex networks. With many different
types of lymphocytes distributing all over the body,
the biological immune system (BIS) [14] can identify
and kill the intrusion antigens. It is seen as a
distributed system. The biological immune network
inspires us to design more advanced defence system
against the collaborative attacks. At first, the immune
network against the attacks determines whether the
strange objects are selfs and detect the attacks [10]. If
they are selfs, the objects are not relative with the
attacks; otherwise, the objects are the nonselfs that
cause the attacks. Detecting the selfs and the attacks
is the first mission of the native immune tier, and
recognizing and classifying the known attacks are the
other responsibilities of the tier. To recognize the
unknown attacks, immune learning and memory are
required for the adaptive immune tier of immune
network [7]. Artificial immune system derives from
BIS was proposed and evaluated on the basis of the
tri-tier immune model and has become an important
research direction in the realm of network security. In
this paper, a cooperative immune model against the
collaborative attacks such as black hole attacks and
wormhole attacks in the mobile ad hoc networks was
proposed and evaluated to detect the attacks and
minimize the attacks.
In this paper, we propose a secure architecture in
MANET based on the framework of the tri-tier
collaborative immune system. The architecture could
apperceive the attack action automatically, and
identify the type of the attack by matching the
immune memory database and immune strategy
database.
The tri-tier immune model consist the native immune
tier, the adaptive immune tier and the parallel immune
tier. The native immune tier is used to detect attacks
in a cooperative way, and the self is the most
important factor in increasing the efficiency and
effectiveness of the attack detection process. Besides,
the native immune tier is also responsible for
recognizing the known attacks. The adaptive immune
tier is used to learn and recognize unknown attacks
cooperatively on the basis of multi agent using RBF
neural networks. The parallel immune tier can be used
to detect the attacks by matching the features of the
7
th

nonselfs. This model has the characteristics of
distribution, robustness and adaptability. Once the
attack is confirmed, our system elaborates a special
responding measure by producing lots of antibodies to
isolate the antigen (invader).
Related Work

The vulnerabilities of the mobile ad hoc networks
have been analyzed in the literature. In the following,
the main characteristics of the vulnerabilities were
reviewed briefly. The collaborative attacks of black
hole and wormhole almost have all the abilities of the
two attacks [3]. For example, the black hole attack
can transmit malicious broadcast information from a
node that the node has the shortest path to the
destination aiming to intercept messages [16].
Wormhole attack can record packets at one location in
the network, tunnel them to other locations, and
retransmit them there into the network [17]. To
defend against the collaborative attacks, a few of
cooperative approaches have been proposed recently.
For example, Cheung decomposed some cyber attacks
into multiple sub attacks and developed a method to
model multistep attack scenarios on the basis of
typical isolated alerts about attack steps . Hidden
Markov models proposed to detect collaborative
attacks [9]. Cuppens made each Intrusion Detection
System (IDS) in some collaborative IDSs [10].
The Intrusion Detection System architecture is
composed of three agents: Monitor agent, decision
agent and killer agent. Monitor agent, which may be
viewed as T-cell, resides on each node and monitors
the neighbor nodes by collecting all packets with its
communication range. Monitor agent encodes
behavior information of its neighbor nodes and send
the codes to decision agent. Decision agents collect
the information from the monitor agent, and make a
judgment by security policies and its immune
memory. They can be viewed as B-cells. When the
decision agents judge that some nodes are invaders,
they will produce killer agents to surround the
invaders and isolate them in the end. The killer agents
can be regarded as antibodies. Other killer agents who
cannot bind with invaders will die after a short time.
The advantage of the architecture is that immunology
is introduced to mobile ad hoc network security
design by using mobile agents as lymphocytes. There
are some limitations in their system. Firstly, each
Monitor agent watches all its neighbors, so each node
is repeatedly monitored by all its neighbor agents.
Secondly, Monitor agent filters, compresses and
encodes the messages from all of the neighbors, and
then sends the codes to Decision agent, which would
burden the network and the Decision agent node.
Thirdly, not detect the unknown attacks. To overcome
the disadvantages of the IDS approaches against the
unknown attacks, the techniques of immune
computation have been investigated for some security
applications, especially in the mobile ad hoc
networks. Artificial Immune System (AIS) to detect
node misbehaviour in a mobile ad hoc network by
using the dynamic source routing protocol [29]. In the
AIS, the threats are the damaged selfs and the foreign
nonselfs, so the threats are the nonselfs in nature. The
vulnerabilities of the AIS depend on its design in
security. RBF is a three-layer neural network which
includes input layer, hidden layer and output layer.
The hidden layer describes its modules to meet with
the needs of different problems. Fig.1 shows the
topology of RBF Neural Network.
Immune system possesses the capabilities of memory,
learning

Fig 1: Topology of RBF Neural Network

as well as self organization and regulation. In
literature [4], by introducing these properties in data
clustering, a data compression algorithm emerged. In
literature [2,6], the above algorithm was used for
selecting data centers in the hidden layers of RBF
network after improvement and simplification, while
the connection weights of the output layers of RBF
network were learned with recursive least square
algorithm. Thus, a new mixed learning algorithm
based on RBF network was formed. This algorithm
does not require the number of data centers in RBF
network to be reassigned.

It is assumed that there are input data in number N,
i.e.
X {
N
}
Wherein,
i=
[
i1,
i2
.,
ip
]
T
R
P

(i=1,..,N)

Represents the input of RBF neural network. A new
data set
C = {c
,..,c
n
}
C
j =
[c
j
c
j
.,.c
nj
]
T
R
P
(j=1,.., h.. h <<
N )

need to be found to determine RBF data centers,
wherein C
represents the clustering of input data, and C is not
necessarily the subset of X. When the principle of
immune system is applied for selecting data centers in
7
th

RBF network, stochastic initialization at certain
number scale is performed on data set C. The
selection process includes the following steps: (1)
with the input data being recognized by all elements
in data set C, these elements compete with one
another. Winners will propagate, i.e., duplicate and
bring about mutations, while the data failing in
recognition will be removed; (2) The elements
recognized by themselves will be removed from data
set C if inhibition occurs due to mutual recognition
between two elements in C; (3) New element is added
in data set C, and the steps (1) and (2) are repeated;(4)
When certain judging condition is satisfied, the
selection process is terminated, and the data centers of
the RBF network model are obtained. In immune
system, the interaction intensity between antigens and
antibodies is represented by the affinity of their
bonds, and antigen-antibody interaction is described
by their similarities. Here the affinity between i
inputted data x
i
and j data center c
j
is denoted by a
ij
,
and the similarity between i data center c
i
and j data
center c
j
is denoted as s
ij
.

III. Architecture
The tri-tier immune model consist of the native
immune tier, the adaptive immune tier and the parallel
immune tier, as shown in Figure 1, new tri-tier
architecture for securing the mobile ad hoc networks.
As the first tier, the native immune tier is used to
detect attacks in a cooperative way, and the self is the
most important factor in increasing the efficiency and
effectiveness of the attack detection process. Besides,
the native immune tier is also responsible for
recognizing the known attacks. The second tier is
adaptive immune tier is used to learn and recognize
unknown attacks cooperatively on the basis of multi
agent using RBF neural networks. First, the native
immune tier detects the selfs, which are defined here
as the normal components of the mobile ad hoc
network. The self model is of the spacetime
properties for the normal states to increase the
precision of self detection, as shown in Figure 2.

Fig 2 Tri-tier Cooperative Immune System with RBF

When the mobile ad hoc network is normal, the
spacetime properties, which identify the self status,
of the normal components are stored into the self
database. The tri-tier immune model is based on the
self model and the self detection because the results
for detecting the selfs in the first step of
immunization can be used to detect more attacks more
quickly than the approach for detecting the attacks
directly. For example, the wormhole attacks attempt
to modify the routing protocol files and the routing
table files of the attacked node so that the wormhole
attacks can transmit their own attacking codes to other
nodes from the compromised node by sending some
attacking packets. Before the wormhole attacks occur,
all the normal nodes of the mobile ad hoc network
store their spacetime properties of each files in the
nodes into a secure self database, and the spacetime
properties of the files can be the absolute pathnames
and the last revision time. Once any file of the core
parts in any node is modified by the wormhole attacks
to change the protocol and the routing table, the self
detection through the self database will return an alert
of detected nonself, that is, the wormhole attacks.
Because the selfs of the normal components are well
known for the system, it is easier to detect whether a
node is a self than to detect whether the node is a
compromised one by unknown attacks by recognizing
the unknown features of the attacks. Moreover, when
the self model is damaged unfortunately, the immune
learning using RBF of the adaptive immune tier and
the parallel immune tier can be used to detect the
attacks by matching the features of the nonselfs, as
shown in Figure 1. Although the compromised node,
whose self model may be damaged by the attacks,
cannot detect the damages with its self model, the
other nodes that are attacked by this node will detect
the attacks with the normal model and the features of
the damaged files in the compromised node.
For instance, when an attack is detected, the
source node of the attack may be a normal one in the
past and now damaged by the attack. This
determination depends on whether the space property
of the source node is already in the self database. If
the search of this space property in the self database
returns nothing, the source node of the attack is sure a
new attacking node. The nonselfs are defined here as
both the damaged components and the new attacking
nodes, which are not acceptable for the
Immune mobile ad hoc network. The nonselfs are
eliminated, and the damaged components are repaired
finally. In the third tier, the parallel immune tier is
used to increase both the efficiency and robustness of
immune computation. In Figure 1, the immune
learning is made by searching the most similar known
attacks in the feature space of attacks.
Native Immune Tier
Self model
Self Detection
Parallel Immune Tier
Cooperative
Agents
Adaptive Immune Tier

Immune Learning
using RBF
7
th

In adaptive immune tier, Detection agent is the core of
the architecture, its immune memory to find the
degree of similarity. Successful matching indicates
that there are invasion behaviors, while not matching
shows that there is no record in immune memory, and
then the upper function will work. Upper function is
used to contrast the codes of specified neighbor which
acts with protocols in strategy library, to identify the
invasion not found in underlying immune memory.
Once invasion is discovered by Detection Agent, it
will send out directions to all neighbor nodes around
the intruder to stop forwarding all intruder messages
to isolate it from the network. Detection Agents are
equivalent to T cells. Their main function is to
identify invaders (antigens), and to organize other
nodes to carry out the attacks against the invaders.
Detection Agents in some nodes are distributed in
different parts of the network. With the nodes are
roaming, they can keep the entire network under
surveillance. Detection Agent should not stay in each
node. It simply uniformly distributes in network. Ad
Hoc network is divided into small zones, and in each
zone, there is one Detection Agent in, which is
responsible for monitoring all nodes in this zone.
Detection Agent is being mapped with RBF neural
network approach as follows in Fig 3

Detection agent terminology of RBF neural
network contains three layer input layer, hidden layer
and output layer input layer x1,x2,xn contains antigen
is simply fan-out, in hidden layer computation work
will be done. Here, the antibodies are present the
weights correspond to cluster centre and the output
function is usually a Gaussian type. The third is
output layer here the weighted sum is calculated.

Fig 3: Detection agent terminology of RBF neural
network.

IV.Characteristics of tri-tier architecture

Cooperative immune systems are applied to
security architecture in the use of Agent technology
for the detection of abnormal behavior and active
response. There are following characteristics in this
architecture:

A) Protection against Collaborative Attack
Collaborative attacks in the mobile ad hoc
networks can cause more damages than individual
attacks. Inspired by the human immune system, a tri-
cooperative immune model is protection against
collaborative attacks in the mobile ad hoc networks.
In Adaptive immune tier, A Multi Agent system has
been adopted in the immune learning to perform
detection in an efficient manner using RBF neural
network. Based on the output, a counter attack agent
is dispatched to surround the invaders and kill it using
the killer agents.

B) Reduce Overhead
Monitor Agent stays in each node to keep all
its neighbor nodes under surveillance. Thats to say,
one node will be repeatedly monitored by all neighbor
Monitor Agents. Monitor Agent sends all message
codes of its neighbor nodes behaviors to Decision
Agent, which will make network overload. Our
Architecture is designed to use one Detection Agent
in one-hop scope to monitor and make decision,
which will decrease greatly the networks load and
node burden.

The performance of the proposed framework will
yield efficient Packet Delivery Ratio, throughput, the
traffic overhead, and the responsiveness of the
system. By experimental, the results will confirm the
effectiveness of the proposed cooperative immune
model in detecting and mitigating these collaborative
attacks from disrupting the protected mobile ad hoc
networks.
V. Conclusion and Future Work

Cooperative immune System was proposed
to defend the ad hoc network under such collaborative
attacks as the black hole attacks and the wormhole
attacks. New tri-tier cooperative immunization-based
framework was designed to detect and immune agent
with RBF neural network approach to identify the
new attack in MANET network. The immune system
and the Agent technology, improves the existing
security model. The system has the characteristics of
distribution, self-adaption, self-learning, and
expandability, and so on.The performance of the
proposed framework will yield efficient Packet
Delivery Ratio, throughput, the traffic overhead, and
the responsiveness of the system. By experimental,
the results will confirm the effectiveness of the
proposed cooperative immune model in detecting and
mitigating these collaborative attacks from disrupting
the protected mobile ad hoc networks.
7
th

For future works, it is interesting to
evaluations of the issues such as tests against other
collaborative attacks, real-time identification of the
selfs, complexity, optimization, and consumption are
also left for future work.

References
[1] Tao Gong and Bharat Bhargava Immunizing
mobile ad hoc networks against collaborative attacks
using cooperative immune model Security and
Communication Networks 2012. 58-68

[2] Xia Ye, Junshan Li, A Security Architecture
Based on Immune Agents for MANET
Proceedings of IEEE ICWCSC CONFERENCE,
IEEE, 2010.

[3] Bhargava B, Zhang Y, Idika N, et al.
Collaborative attacks in WiMAX networks. Security
and Communication Networks 2009; 2(5): 373391.

[4] Y. Zhang and W. Lee, Intrusion Detection in
Wireless Ad- Hoc Networks, Proceedings of the 6th
Annual International Conference on Mobile
Computing and Networking, MobiCom 2000, pp.
275-283.

[5] Ping Yi, Yiping Zhong, Shiyong Zhang, An
immunity-based security architecture for mobile Ad
Hoc networks, Journal of Electronics (China),
Vol.23 No.3, May 2006.

[6] Wang Yinglong, Wang Jizhi, Wang Merqin, Gao
Zhenming, An analysis technology of security
routing protocol for MANET, Computer Engineering
and Application (china). Vol.23 December 2005.

[7] S. Sathyanath, F. Sahin, "AISIMAM An AIS
based intelligent multi agent model and its application
to mine detection problem," In: Proceedings of the 1st
International Conference on Artificial Immune
Systems, Canterbury, UK, 2002, pp. 9 11.

[8] Ramaswamy S, Fu H, Nygard K. Effect of
cooperative blackhole attack on mobile ad hoc
networks. In ICWN. 2005.

[9] Sarafijanovic S, Le Boudec JY. An artificial
immune system approach with secondary response for
misbehaviour detection in mobile ad hoc networks.
IEEE Transactions on Neural Networks 2005; 16(5):
10761087.

[10] Yu-Sung W, Foo B, Mei Y, Bagchi S.
Collaborative intrusion detection system (CIDS): a
framework for accurate and efficient IDS. In
Proceedings of Computer Security Applications
Conference. 2003; 234244.

[11] Ying Zhou, De-ling Zheng, Zhi-Liang Qiu, Guo-
ya dong, "The application of RBF networks based on
artificial Immune algorithm in the performance
prediction of steel bars," Journal of University of
Science and Technology Beijing, 2005, pp.23-26.

[12] Li Changrong, Qu Zuyu, Yang Dejun,
"Clustering analysis of corrosive factors of soil,"
Journal of Chinese Society for Corrosion and
Protection, 1994, pp. 138-143.

[13] A.K.Ghosh and A.Schwartzbard.A study in
using neural networks for anamoly and misuse
detection, Proceedings of The 8
th
USENIX security
symposium, 1999.

[14] Ziv Y, Ron N, Butovsky O, et al. Immune cells
contribute to the maintenance of neurogenesis and
spatial learning abilities in adulthood. Nature
Neuroscience 2006; 9: 268275.

[15] R. Heady, G. Luger, A. Maccabe, and M.
ServillaThe Architecture of a network level intrusion
detection system. Technical report, Computer
Science Department, University
Of New Mexico, August 1990.

[16] L. Zhou and Z. J. Haas. Securing adhoc
networks. EEE Network, 13(6):24{30, Nov/Dec
1999.

[17] Lakshmi Venkatraman. Secured routing
protocol for ad-hoc networks. Master's thesis,
University of Cincinnati, OH,
March 2000.

Various Papers in Our College

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Various Papers in Our College

Încărcat de

Drepturi de autor:

Formate disponibile

Proceedings of the Seventh NCVIT

S-ar putea să vă placă și