PERFORMANCE EVALUATION OF ENERGY EFFICIENT CLUSTERING PROTOCOL FOR CLUSTER HEAD SELECTION IN WIRELESS SENSOR NETWORK

International Journal of Peer to Peer Networks (IJP2P) Vol.5, No.3, August 2014
DOI : 10.5121/ijp2p.2014.5301 1
PERFORMANCE EVALUATION OF ENERGY
EFFICIENT CLUSTERING PROTOCOL FOR
CLUSTER HEAD SELECTION IN WIRELESS
SENSOR NETWORK
Harpreet Kaur1
and Prof Ameeta Seehra 2
1
Department of Electronics and Communication Engineering, GNDEC, Ludhiana, India
2
Department of Electronics and Communication Engineering, GNDEC, Ludhiana, India
ABSTRACT
Wireless sensor network are emerging in various fields like environmental monitoring, mining, surveillance
system, medical monitoring. LEACH protocol is one of the predominantly used clustering routing protocols
in wireless sensor networks. In Leach each node has equal chance to become a cluster head which make
the energy dissipated of each node be moderately balanced. We have pioneered an improved algorithm
named as Novel Leach based on Leach protocol. The proposed algorithm shows the significant
improvement in network lifetime .Comparison of proposed algorithm is done with basic leach in terms of
network life time, cluster head selection, energy consumption, and data transmission to base station. The
simulation results shows that proposed algorithm can reduce network energy consumption and prolong
network life commendably. Simulation of our protocol is done with Matlab.
Keywords
Wireless sensor network, Leach, Leach C, Leach CC, Cluster Head Selection, Network life time, residual
energy.
1. INTRODUCTION
Wireless Sensor Networks are usually self-organized wireless ad hoc networks comprising of a
large number of resource constrained sensor nodes. The vital role of these sensors nodes is
systematic collection of data. After that sensor nodes transmit aggregated data to BS. Wireless
sensor network has spacious application in military recce, medical assistance, agriculture,
environmental monitoring and other infomercial areas.Wireless Sensor Network [1] produce large
number of data which is aggregated at numerous levels. A multidimensional aggregation
approach [13] is considered for exhibiting the node parameters for each network. The
performance of sensor network is determined by memory, battery power, bandwidth etc.
Life span of sensor network is an important parameter for efficient design of data aggregating
schemes in sensor network. Energy organization plays important role in the deployment of sensor
networks. The available energy is considered a major factor when designing protocols in a sensor
network. Hence existing energy plays important role in designing of protocol in wireless sensor
network.Low-Energy Adaptive Clustering Hierarchy (LEACH) is a clustering routing algorithm
in WSN.In LEACH cluster head is determined randomely,it is not based on remaining energy of

2
sensor nodes and which results that less energy nodes come to an end of energy in advance. It
decrease in life time of sensor network. Our research analysis the significance of Leach protocol
in cluster head selection and energy saving of sensor network.
2. INTRODUCTION TO LEACH PROTOCOL
W. R Heinzelman proposed LEACH protocol, which based on cluster structure and hierarchical
technology. Relative to the traditional protocol, LEACH could save a greater degree of energy.
LEACH is the most commonly used clustering algorithm in wireless sensor network .In LEACH,
the nodes are deployed in specific area and organized into clusters. The communication process is
divided into rounds .Each round having two phases namely set-up and steady-state phases. Each
cluster has a cluster head which is accountable for creating a TDMA (time division multiple
access) table. It intimate its member sensor nodes when to get transmit data packets The main
function of cluster head is to collect data which is coming from different sensor member nodes
and transmit data packets to BS.Cluster head serve as a relay node to provide assistance to senor
nodes to shorten the transmission distance also it saves energy.
2.1 Setup Phase
Each node in the monitoring field decides self-governing of other sensor nodes whether selected
node can be cluster head in current round .In this phase, each sensor node generate a random
number between 0 and 1.If the number is larger than threshold value then node become a cluster
head for current round and send availability message to its neighboring sensor nodes .In the set-
up phase, the node becomes a cluster head with a probability P. and send its data packets. On the
other side, the regular nodes choose their cluster-head based on the least communication energy
to reach the cluster-head. The threshold is set as
ܶሺ݊ሻ = ൞
‫݌‬
1 − ‫݌‬ ∗ ቀ‫ ݀݋݉ ݎ‬
1
‫݌‬ቁ
0

݂݅ ݊ ߳ ɢ
݂݅ ݊ ̸߳ ɢ
Where P denotes the percentage of cluster head, r denotes the current round and G denotes as set
of nodes that are not become cluster heads. According to this threshold value, each node in the
sensor network have chance to become a cluster head at 1/P round. When the first round occur
each node has probability to become a cluster head and the nodes which are chosen to become
cluster head in first round, they can’t participate for cluster head selection in next round Once the
cluster-heads have been chosen, the cluster-heads use CSMA MAC protocol to broadcast
advertisement messages to the remaining sensor nodes. The regular nodes must keep their
receivers on during this phase to hear the advertisements of all the cluster-heads. After this phase,
each regular node decides which cluster to join for the current round. The other sensor node will
inform the cluster-head that it will become a member of the cluster. Each regular node transmits
this information back to the cluster-head again using a CSMA MAC protocol. The cluster-head
receives all the messages for sensor nodes and would interested to intersect the clusters. Based on
the number of regular nodes in the cluster, the cluster-head creates a TDMA schedule telling each
regular node when it can transmit. This schedule is broadcast back to the regular nodes in the
cluster.

3
2.2 Steady State Phase
The cluster head selection in each cluster during the setup phase ensures the data transmission in
steady state phase when there is need to transmit data, first they will send data to Ch. during idle
slot canned in TDMA schedule table. Cluster heads should maintain communication status at all
times and it will help to receive data from different sensor nodes. After receiving all the data sent
by their members, CHs will aggregate them firstly and then send them to BS. When some sensor
nodes may detect similar data them data reduce unwanted bandwidth cost, energy consumption is
less. To overcome the problem of Cluster head dies early, a new round starts and new clusters
formation will be done in the sensor network.
3. SYSTEM MODEL
3.1 A Radio Energy Model
We consider a wireless sensor network which consist of 100 sensor nodes, area used is
100x100 square meters.
Figure 1 Radio energy model
We use a simplified model shown in figure 1 for the communication energy dissipation. In
accordance with the distance between the source node and the destination node, we used the free
space (d2
power loss) and the multi-path fading (d4
power loss) channel models. Thus, the energy
spent for transmission of an l-bit packet over distanced d is:
ETX (l,d) = E TX_elec *l +Eamp (d) *l (1)
L Eelec + l Ɛ FS *d 2
d˂ d crossover (2)
ETX (l,d ) = L Eelec + l ƐTR *d4
d ≥ do (3)
Eelec is the energy being dissipated to run the transmitter or receiver circuitry. The parameters
Ɛ TR and Ɛ FS is the amount of energy dissipates per bit in the radio frequency amplifier according
to the crossover distance d0,

4
do = ට
Ɛ೑ೞ
Ɛ ౪ೝ
(4)
to receive l message bit , the radio apply energy
ERX (l) = ERX_elec(l) = l E elec (5)
3.2 Optimal Clustering
In WSN the optimal probability of any node to become a cluster head is very important. In
optimal clustering energy consumption is properly distributed along with all sensor nodes In this
radio model area is assumed A =M X M square meter. Let there be n Nodes are randomly
distributed .Assume that base station is located on fix place.
If there are K cluster in sensor network, then average node per cluster is noticed as n/k.. So there
will be one cluster head and remaining (n / k -1) are non-cluster head nodes.
.ENON_ch =L.EELEC +EAMP (L, d) (6)
Now energy dissipated by cluster head is given by
ECH =L EELEC (n /k -1) +L EDA n / k +L EELEC +L EAMP d4
to BS (7)
Where k is the number of Cluster, ,EDA is the data aggregation and dto BS is the average distance
between a CH and BS which is given by
d2
BS = ʃඥ‫ݔ‬ଶ + ‫ݕ‬ଶ ܺ
ଵ
஺
= 0.765 ܺ
ெ
ଶ
(8)
Hence, energy dissipated by a non CH node is given by
ENON CH = L × ( EELEC + Ɛ FS x d2
CH ) (9)
Where d CH is the average distance between a non CH node and its associated CH, which is given
by equation
d2
CH = ʃ ʃ ( x2
+ y2
) x ρ ( x , y) dx dy =
ெ
ଶగ௞
ଶ
(10)
Where ρ ( x , y) is the node distribution and M is the area of monitoring field.
Now, the total energy dissipated is
ET = E CH +E NON CH (11)
Now,
ET = L × (2 n × EELEC + n × EDA + Ɛ F S × (k × d2
to BS + n ×
ெ
ଶగ௞
ଶ
ሻ (12)

5
Now we can find the minimum no of cluster by setting the derivative of ET with respect to 0
K opt = ට
௡
ଶగ
× ට
ఌ೑ ೞ
ఌ೘ ೛
×
ெ
ௗమ
್ೞ
(13)
Average energy is estimated as
Eaverage =
ଵ
ே
× ‫ܧ‬் × ሺ1 −
௥
ோ
ሻ (14)
Where R denotes the total rounds of network lifetime whereas r denotes the round of sensor
network.
4 SIMULATION ANALYSIS
To evaluate the performance our proposed protocol we use matlab .
4.1. Network Model
1. The Base station is immovable at far away from the sensor nodes
2. The sensor nodes are homogeneous and energy constrained with even energy.
3. The energy rate for transmitting a data packet depends on the distance of transmission.
4. The communication channel is symmetric.
4.2 .Parameters For Simulation
We simulate leach, leach c, leach cc, proposed algorithm for wireless sensor network. The basic
parameters are listed in Table 1.
Table 1 Summary of the parameters
S.NO PARAMETERS VALUE
1 No of Nodes 100
2 Network Size (100 x 100) m2
3 Initial Energy(Eo) 2 J
4 Energy for Transmission (ETX) 50nJ/bit
5 For Energy Reception (ERX) 50nJ/bit
6 Free Space Energy Loss (Efs) 10 pJ / bit /m2
7 Multi path Energy Loss (Emp) 0.0013pJ / bit /m4
8 Maximum no. of Rounds (rmax) 700
9 Location of BS(Sink) (50,175)
10 Adaptive Energy(EDA) 5n J/bit

Fig
Figure shows the deployment of sensor
100x 100 meter square. Base station is located at (50,175)
5. FUZZY LOGIC CONTROL
In Fuzzy inference system, the rules are
called as linguistic variable, Rules are simply
rule is called antecedent and THEN
are AND, OR etc. The combined truth of the predicate is determined by implication rules such as
MIN-MAX ( Zadeh) and bounded arithmetic sum.
In FIS, there are four important steps as shown below:
1. Fuzzification: It convert the system inputs which are crisp values into fuzzy sets.
2. Rule evaluation: It fuzzified inputs
rules.
3. Aggregate conclusions: It conclude the
4. Defuzzication: It convert the fuzzy set obta
value.
The input and output variable used in our work as stated as below;
Figure 2 Node deployment of sensor nodes
shows the deployment of sensor nodes. All nodes are deployed randomly in the area of
station is located at (50,175).
CONTROL
the rules are generated based on input variable and output variable
Rules are simply based on IF-THEN technique. The IF part of fuzzy
and THEN part is known as consequent .Fuzzy implication operators
The combined truth of the predicate is determined by implication rules such as
bounded arithmetic sum.
In FIS, there are four important steps as shown below:
the system inputs which are crisp values into fuzzy sets.
fuzzified inputs variables and evaluate them to the antecedents
conclude the outputs of all rules.
the fuzzy set obtained by the inference system into a single crisp
The input and output variable used in our work as stated as below;
6
in the area of
on input variable and output variable,
The IF part of fuzzy
consequent .Fuzzy implication operators
The combined truth of the predicate is determined by implication rules such as
evaluate them to the antecedents of the fuzzy
a single crisp

1. Node Energy: It shows the
variable energy. Linguistic values are
2. Distance to BS: It is the distance between node and base station .The linguistic variable for
Distance to BS is far,medium ,close.
3. Chance: it is the chance to become a cluster head .The linguistic variable are us
weak , low weak, weak ,low medium, high medium, medium, low strong, strong, very strong.
The linguistic variable used to represent th
high,medium,low for energy as in fig
figure 4.The output of node to become a cluster head is represented by chance is divided into nine
levels are very weak ,low weak, wea
very strong.
We have used 32
= 9 rules for fuzzy base rule.
corresponding linguistic states are shown in following
Figure
Figure 4 Membership function plot for Distance to BS
It shows the energy level available in each node, designated by
values are low, medium, high.
It is the distance between node and base station .The linguistic variable for
o BS is far,medium ,close.
it is the chance to become a cluster head .The linguistic variable are us
k ,low medium, high medium, medium, low strong, strong, very strong.
The linguistic variable used to represent the node energy and distance are divided into three levels
high,medium,low for energy as in figure 3; and close ,medium ,far for distance to BS as in
4.The output of node to become a cluster head is represented by chance is divided into nine
are very weak ,low weak, weak ,low medium, high medium, medium, low strong, strong,
for fuzzy base rule. The membership function developed and their
corresponding linguistic states are shown in following figures.
ure 3 Membership function plot for Energy
4 Membership function plot for Distance to BS
7
energy level available in each node, designated by the fuzzy
It is the distance between node and base station .The linguistic variable for
it is the chance to become a cluster head .The linguistic variable are used are very
k ,low medium, high medium, medium, low strong, strong, very strong.
e node energy and distance are divided into three levels
3; and close ,medium ,far for distance to BS as in
4.The output of node to become a cluster head is represented by chance is divided into nine
k ,low medium, high medium, medium, low strong, strong,
The membership function developed and their

Figure 5 Surface diagram of Energy, Distance to BS, chance
S.NO ENERGY
1. LOW
2. LOW
3. LOW
4. MEDIUM
5. MEDIUM
6. MEDIUM
7. HIGH
8. HIGH
9. HIGH
Surface diagram of Energy, Distance to BS, chance
TABLE 2. FUZZY RULES
ENERGY DISTANCE TO BS CHANCE
FAR V WEAK
MEDIUM L WEAK
CLOSE WEAK
FAR L MEDIUM
MEDIUM H MEDIUM
CLOSE MEDIUM
FAR L STRONG
MEDIUM STRONG
CLOSE V STRONG
8

9
6. RESULTS AND DISCUSSIONS
Figure 6. Comparative analysis of alive node
Table 3. Simulation results shows the no of alive node.
SNO ALGORITHM First node
dead
HND % Improvement
FND
% Improvement
HND
1. LEACH 28 43 460.7 553.4
2. LEACH C 148 259 6.08 8.49
3. LEACH CC 28 37 460.7 659.45
4. NOVEL 157 281 560.7 553.488
A comparison is drawn between the performance of proposed leach with LEACH, LEACH C,
LEACH CC Protocol in terms of network lifetime and in terms of FND, HND.Performance.
Comparison results for network lifetime of proposed leach with that Leach,leach c,leach cc
protocol shown in figure 6.The network lifetime of proposed algorithm has improved further from
other existing protocols. Proposed algorithm improves the network life time by 84.69 %
0 100 200 300 400 500 600 700
0
10
20
30
40
50
60
70
80
90
100
Alive Node in WSN
No of Rounds (seconds)
NoofAliveNodes NOVEL Leach
LeachC
LeachCC
Leach

10
.
Figure 7. Comparative analysis of packets transmitted to BS
Table 4 PACKET TRANSMITTED TO BS
SNO ALGORITHM PACKETS SEND TO BS
1. LEACH 69,100
2. LEACH C 82,200
s3. LEACH CC 1,24,736
4. NOVEL 2,83,076
The comparative analysis of packets transmitted to Base Station shows that proposed algorithm
transmit more data packets to Base Station .The comparative analysis of Novel leach is done with
Leach, leach c, leach cc respectively.
0 100 200 300 400 500 600 700
0
0.5
1
1.5
2
2.5
3
x 10
5 Packets Transmitted in WSN
NoofPacketsreceivedatBaseStation
Novel Leach
LeachC
LeachCC
Leach

11
Figure 8.Comparative analysis of Energy Consumption
Table 5 Energy Consumption Results at 700 rounds
S no Protocol Name Total energy remaining
1. LEACH 0.3587 joule/second
2. LEACH C 0.7938 joule/second
3. LEACH CC 3.8920 joule/second
4. NOVEL LEACH 8.2194 joule/second
The comparative analysis of energy consumption in the sensor network. Leach has very less
remaining energy as compare to Novel LEACH. Novel Leach consume less energy as compare to
Leach c,leach cc.
7. CONCLUSION
To increase Network life time is a challenging issue in WSN .In our proposed protocol, We use
two parameter energy, distance to Base Station for cluster head selection. The proposed
0 100 200 300 400 500 600 700
0
10
20
30
40
50
60
70
80
90
100
Energy consumption in WSN
TotalEnergy(Joule)
Novel Leach
LeachC
LeachCC
Leach

12
algorithm results conclude that it is better than Leach, Leach C, and Leach CC. It prolongs
network life time by 84 % and less energy consumption as compare to basic LEACH.
REFERENCES
[1] Adrian Perrig , John Stankovic and David Wagner,” Security in wireless sensor network”,. Vol. 47,
pp.6-9 ,June 2004.
[2] Bolian Yin, Hongchi Shi, and Yi Shang ,”Analysis of Energy Consumption in Clustered Wireless
sensor Networks” IEEE Transactions on Mobile Computing, vol. 3, pp. 272–285, July 2007
[3] Curt Schurgers and Mani B. Srivastava “Energy efficient routing in Wireless Sensor Networks”,
IEEE Personal Communications Mag., Vol.7, No.5, pp.16-27, Oct. 2000
[4] Fan Xiangning and Song Yulin, “Improvement on LEACH Protocol of Wireless Sensor Network”,
IEEE Communications Magazine,, 40(8), pp.102-114, Mar.2002.
[5] Hiren Kumar Deva Sarma and Avijit Kar, “Security Threats in Wireless Sensor Networks” IEEE
Communications Magazine, 42(4): 122-134, August 2002.
[6] Ioan Raicu ,”Routing Algorithms for Wireless Sensor Networks”, IEEE Wireless Communications,
vol.46, pp.110-119,Sept.2002.
[7] Jan Steffan ,Ludger Fiege, Mariano Cilia and Alejandro Buchmann, “Scoping in Wireless Sensor
Networks”, IEEE Network, 18(1), pp.15–21, May 2004.
[8] John Buckley, Kevin Aherne and Cian O'Mathuna, “Antenna Performance Measurements Using
Wireless Sensor Networks”, Proceedings of the IEEE, VOL. 91, pp. 1247-1256, August 2003.
[9] Kay Romer and Friedemann Mattern, Eth Zurich,”The design space of wireless sensor network”,
CACM, vol. 43, pp. 74–82,Mar. 2000.
[10] Kun Zhang and Cong Wang, “A Secure Routing Protocol for Cluster-Based Wireless Sensor
Networks Using Group Key Management,” IEEE Xplore,vol.37, pp. 178-185, jan.-2008.
[11] L.Li, and J.Y. Halpern, “Minimum-Energy Mobile Wireless Networks Revisited,”, IEEE International
Conference on Communications (ICC). Vol. 1, pp. 278-283, 2001.
[12] L. Subramanian and R. H. Katz “An Architecture for Building Self Configurable Systems”, in the
Proceedings of IEEE/ACM Workshop on Mobile Ad Hoc Networking and Computing, Boston, MA,
August 2000.
[13] M. Ibrahim Channa and Irum Memon, “Real Time Traffic Support in Wireless Sensor Networks”
IEEE Comm. Magazine, pp.119-133, June 2001.
[14] Mihaela Cardei , Jie Wu and Mohammad O. Pervaiz,”Maximum Network Lifetime in Wireless
Sensor Networks with Adjustable Sensing Ranges”, IEEE Communications Magazine, pp 102-114,
Aug.2002.
[15] Peng-Jun Wan, Member and Chih-Wei Yi, “Coverage by Randomly Deployed Wireless Sensor
Networks”, IEEE Wireless Communications and Networking Conference, pp.16–20, Mar. 2003.
[16] Q. Li and J. Aslam and D. Rus, “HPAR: Hierarchical Power-aware Routing in Sensor Networks”, In
Proceedings of the DIMACS Workshop on Pervasive Networking, May, 2001.
[17] Qi YAO, Seng- Kee TAN, Yu GE, Boon-Sain YEO and Qinghe YIN, “An Area Localization Scheme
for Large W ireless Sensor Networks” MOBICOM, pp269- 286, July 2001.
[18] Sang-Sik Kim and Ae-Soon Park, ”Mobility Support for Users in Wireless Sensor Networks”, IEEE
Communications Magazine, vol. 40, pp.124-156 Aug. 2002
[19] Shafiq Hashmi and Hussien T. Moufth, “A New Transport Layer Sensor network protocol”, IEEE
Transactions, vol. 5,pp.118-156,Mar.2003
[20] Shangwei Duan and Xiaobu Yuan, ”Exploring Hierarchy Architecture for Wireless Sensor Networks
Management”, IEEE Communications Magazine, 40(8), pp.102-114, Aug. 2002.
[21] Rogaia Mhemed et al.,”An Energy Efficient Fuzzy Logic Cluster Formation Protocol in wireless
Sensor Networks,” The 3rd International Conference on Ambient Systems, Networks and
Technologies,Niagara Falls, Ontario, Canada. 27-29 August 2012.
http://nrl.northumbria.ac.uk/7042/

13
Authors
Harpreet Kaur is the mtech student in Department of Electronics and Communication
Engineering , Guru Nanak Dev Engineering College Ludhiana. She has published several
research papers in wireless communication .Her area of research interest is Wireless
communication systems.
Prof. Ameeta Seehra is presently working as associate professor in Department of
Electronics and communication engineering , Guru Nanak Dev Engineering college
,Ludhiana. She has received her B.E. in Electronics and Comm. engg in 1986 and M.Tech.
Electronics and Comm. engg in 1997.She has guided more than 20 students for Mtech
Thesis .She is a member of ISTE. Her area of research interest is advanced
communications.

PERFORMANCE EVALUATION OF ENERGY EFFICIENT CLUSTERING PROTOCOL FOR CLUSTER HEAD SELECTION IN WIRELESS SENSOR NETWORK

Recommended

More Related Content

What's hot (14)

Similar to PERFORMANCE EVALUATION OF ENERGY EFFICIENT CLUSTERING PROTOCOL FOR CLUSTER HEAD SELECTION IN WIRELESS SENSOR NETWORK (20)

More from ijp2p (20)

Recently uploaded (15)

PERFORMANCE EVALUATION OF ENERGY EFFICIENT CLUSTERING PROTOCOL FOR CLUSTER HEAD SELECTION IN WIRELESS SENSOR NETWORK