Energy Efficient Clustering Protocol for Wireless Sensor Networks using Particle Swarm Optimization Approach

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 07 | July -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 838
Energy Efficient Clustering Protocol for Wireless Sensor Networks
Using Particle Swarm Optimization Approach
Ms. Divya Garg1, Dr. Pardeep Kumar2, Ms. Kompal3
1M.Tech Scholar, Dept. of Computer Science & Applications, Kurukshetra University, Haryana, India
2Associate Professor, Dept. of Computer Science & Applications, Kurukshetra University, Haryana, India
3Assistant Professor, Dept. of Computer Science, Govt. College Chhachhrauli (Yamunanagar), Haryana, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - WSN comprises of a considerable amount of
small and limited power sensor elements that are arbitrarily
or physically conveyed over an unattended target region.
WSNs have potential applications in atmosphere observing,
calamity cautioning frameworks, medicinal services, security
surveillance, and reconnaissance frameworks. The main
drawback of the wireless sensor network is the restricted
power sources of the sensing elements. Expanding the lifetime
of the Wireless Sensor systems, energy preservation measures
are vital for enhancing the execution of WSNs. This paper
proposes LEACH-P which is a novel approach to improve
existing LEACH protocol using PSO based clustering. The
proposed algorithm is simulated broadly and the results are
compared with the existing algorithm to determine its
supremacy in terms of network lifetime, stability period and
number of data transmitted to the base station.
Key Words: Wireless Sensor Networks (WSN), gateways,
Cluster Head (CH), Particle Swarm Optimization.
1. INTRODUCTION
1.1 Background
Wireless Sensor Networks are having large network in
which huge amount of sensor nodes are present that are
forming a network with their self-organizing property. The
variety of applications includes health care, military, critical
infrastructureprotection(Akyildiz,Su,Sankarasubramaniam,
& Cayirci, 2002), and non-military personnel (e.g., disaster
management). In WSN the small sensor nodes are
categorized by restricted processing power sources. In this
manner, energy preservation of the sensors is the most
demanding concern for the long run process of WSNs.
Numerous issues have been contemplated for this reason
that include low-power radio communication equipment
(Calhoun, 2005),powerawaremediumaccesscontrol (MAC)
layer conventions (Ahmad, 2012) (Aykut, 2011) and so on.
Therefore, energy efficient clustering and routing
procedures (Abbasi & Mohamad, 2007) (Kemal & Younis,
2005) are the most encouraging regions that have been
contemplated widely in such manner.
In a two-level WSN, sensing elementsarepartitionedinto
a few groups which are known as clusters. Everygrouphas
a pioneer called cluster head (CH). Every one of the sensing
element sense neighborhood information and transmit it to
their relating CH. At that point, the cluster heads combine
the local information and then transmit it to the base station
(BS) specifically or by means of different CHs. A cluster
based model of wireless sensor network is appeared in Fig.
1. Clustering sensors has various advantages which are as
follows: (1) It empowers information total at CH to dispose
of the repetitive and uncorrelated information;
consequently, it saves power of the sensing elements. (2)
Routing can be all the more effectively achieved on the
grounds that only CHs need to keep up the nearby path set
up of different CHs and subsequently require little steering
data; this in turn enhances the adaptability of the system
essentially. (3) It preserves correspondence transfer speed
as the sensor nodes communicate with their CHs only and
therefore stay away from trade of excessinformationamong
them.
Though, CHs tolerate some additional work load
contributed by their cluster members as they collect the
detected information from their group member sensors,
combines them and convey it to the BS. In addition, in
numerous WSNs, the CHs are typically chosen among the
ordinary sensor nodes which can expire rapidly for this
additional work load. In this remarkable circumstance,
numerous scholars (Gupta & Younis, 2003) (Low, 2008)
(Kuila & Jana, Improved load balanced clustering algorithm
for wireless sensor networks., 2012) (Kuila, Gupta, & Jana, A
novel evolutionary approach for load balanced clustering
problem for wireless sensor networks., 2013) (Bari, Wazed,
Jaekal, & Bandyopadhyay, 2009) have proposed theusage of
some extraordinarysensing elementscalledgateways,which
are provisioned with additional power. These gateways
demonstrates like cluster heads and are in charge of a
Base
station
Internet
Sensor
node
Cluster
head
. .Fig-1: A Wireless Sensor Network Model

similar function of the CHs. Along these lines, gateways and
CHs are utilized reciprocally in the rest of the paper.
Tragically, the gatewaysarelikewisebattery-workedand
consequently energy constrained.Lifetimeofthegateways is
exceptionally vital for the long run operation of the system.
The transmission energy (E) related with distance(d)on the
basis of following formula, i.e. E d λ ,where λ is the path
loss exponent and 2 λ 4 (Habib & Sajal, 2008) . In this
way, minimization of transmission separation can decrease
the energy utilization. In any case, a few applications are
extremely time-basic in nature. Subsequently, they ought to
fulfill strict delay constraints so that the BS can get the
detected information within a predeterminedtimebound.In
any case, the delay is relative to the quantity of forwards on
the dissemination path between a source and the sink. With
a specific end goal to limit the delay, it is important to limit
the quantity of forwards, which can be accomplished by
boosting theseparationbetweencontinuousforwards. Along
these lines, while outlining routing procedures we have to
fuse an exchange off between transmission separation and
quantity of forwards as they stance two clashing
destinations. Moreover, load balancing is another critical
issue for WSN clustering. Especially, this is a problem that
needs to be addressed when the sensor nodes are not
disseminated consistently. In this paper we address the
accompanying issues:
 Energy proficient routing with an exchange off
between transmission distance and number of
information forwards.
 Energy proficient load balanced grouping with
energy preservation of the WSN.
Keeping in mind the end goal to acquire a quicker and
proficient arrangement of the clustering and routing issue
with the above issues, a metaheuristic approach, for
example, particle swarm improvement (PSO) is extremely
attractive. The primary goal of this paper is to enhance
existing LEACH with an effective PSO-based clustering for
WSNs with the thought of energy utilization of the sensor
hubs for extending system life time.
1.2 Authors' Contribution
In this paper, an energy efficient protocol for WSN is
proposed using particle swarm optimization. The proposed
research work will execute the PSO with leach protocol. The
PSO in clustering for optimal selection of cluster head is
applied to enhance the advancement in the residual energy
of node by sending a data packet to the cluster head which is
located very nearest to the Base station. The proposed
LEACH (LEACH-P) discovers a path from all the gateways to
the base station which has comparably lower overall
distance with less number of data forwards.
The proposed LEACH takes care of energy consumption
of the normal sensor nodes as well as the gateways. We
perform extensive simulation on the proposed methods and
evaluate them with several performance metrics including
network life-time, stability, energy consumption and total
number of data transmitted. The results are compared with
LEACH (Heinzelman, Chandrakasan, & Balakrishnan, 2002)
which is a widespread cluster-based routing procedure. Our
key contributions can be brief as follows:
 Sensor nodes are deployed arbitrarily along with
a few gateways.
 The PSO in clustering for optimal election of
cluster head is applied fortheadvancementinthe
residual energy of node by sending a data packet
to the cluster head.
 Simulation of the proposed procedure to
determine supremacy over existing procedure.
The rest of the paper is organized as follows. The
related work is discussed in Section 2. LEACH isdefinedin
Section 3. An outline of particle swarm optimization is
given in Section 4. The system model is presented in
Section 5 which comprises energy model and network
model. The proposed algorithm i.e. LEACH-P is described
in Section 6. The experimental Setup and the simulation
results are described in Sections 7 and 8 respectively and
we conclude in Section 9.
2. RELATED WORK
(Heinzelman, Chandrakasan, & Balakrishnan, 2002)
have been proposed a mainstream cluster-based steering
technique LEACH which is able to control the cluster head
work load among the sensing elements that is profitable for
balancing the load. However, the fundamental weakness of
this method is that a sensing element with low power might
be chosen as a CH which may die hastily. Furthermore, the
CHs communicate with a base station by means of one-hop
which is unrealistic for WSNs with extensive scope region.
Therefore, a lot of procedures have been established to
advance LEACH which can be found in (Tyagi & Kumar,
2013)], (Al-Refai, 2011). (Kuila & Jana, An energy balanced
distributed clustering and routing algorithm for wireless
sensor networks, 2012) have been proposed a cost- based
disseminated energystableclusteringandroutingprocedure
for selection of cluster head and cluster creation. But, the
procedure experiences the network issue of the chose CHs.
(Bari, Wazed, Jaekal, & Bandyopadhyay, 2009) proposed
a GA-based methodology for information directing between
gateways in a two-tire wireless sensor network. The relay
sensing elements may shape a system among them to route
information towards the base station. In this design, the
lifespan of a system is determined mainly by the lifespan of
these relay sensors. An energy-aware communication
strategy can greatly extend the lifetime of such networks.
However, integer linear program (ILP) formulations for
optimal, energy-aware routing rapidly become
computationally intractable andarenotsuitableforconcrete
systems.

(Kennedy & Eberhart, 1995) proposed a system for
development of reliable nonlinear limits. The system was
found through re-enactment of an enhanced social model;
thusly the social analogy is discussed, however the
calculation stays without metaphorical support. The paper
portrays the particle swarm improvement idea as far as its
forerunners, rapidly investigating into the phases of its
advancement from social rebuilding to analyzer. The
improvement of a couple of perfect models is laidout,andan
execution of one of the guidelines is inspected. The
relationship between particle swarm development andboth
manufactured life and hereditary controls are depicted.
(Singh & Lobiyal, 2012) proposed an Energy-efficient
cluster head selection using Particle Swarm Optimization
approach and examination of Packet Retransmissions in
WSN. Particle Swarm Optimization (PSO) method is useful
for creating vitality mindful clusters with an ideal choice of
the group head. The PSO eventually diminishes the cost of
discovering ideal position for the cluster head hubs. The
execution of PSO approach is achieved inside the cluster
instead of the sink where a user is able to extract data,which
makes it a semi-distributed method. The choice criteria of
the target capacity depend on the lasting vitality, least
normal distance from the member hubs and head count of
the likely head hubs. (Singh & Lobiyal, 2012)] and (Abdul,
2007) have utilized the PSO for CH choice among the typical
sensor hubs and don't deal with the groupdevelopment.PSO
and ant colony optimization (ACO) are utilized as a part of
WSNs for other optimization issues likewise and they can be
found in (Saleem, 2011), (Kulkarni, 2011), (Zungeru, 2012).
(Kuila & Jana, 2014) have proposed Energy effective
clustering and routing which are two surely understood
advancement issues which have been studied broadly to
increase lifespan of wireless sensor networks (WSNs) and
describes Linear/ Nonlinear Programming (LP/NLP)
definitions of these issues taken after by two proposed
procedures for the same in view of particle swarm
optimization (PSO). The routing procedure is established
with a well-organized particle encoding scheme and multi-
objective fitness function. The clustering algorithm is
described by considering power maintenance of the nodes
through load balancing.
(singh, 2016) have proposed a Multi-HopLEACHwhichis
vitality efficient routing convention for wireless sensor
network; a strategy of routing is recommended in view of
both particle swarm optimization technique and V-LEACH
protocol and multi-hop system design. Multi hop
communication has been applied to limit vitality dissipation
in the transmission from cluster head to the sink.
3. LEACH: LOW-ENERGY ADAPTIVE CLUSTERING
HIERARCHY
LEACH is a self-organizing,adaptiveclustering procedure
that uses randomization to circulate the energy load
uniformly among the sensors in the system. In LEACH, the
sensors sort out themselves into nearby formed clusters,
with one element acting as the local base station (BS) or
cluster-head. LEACH comprises arbitrary circulation of the
high-powered cluster-head position such that it circulates
among the numerous sensors in order not to drain the
battery of a single senor. Moreover, LEACH achieves
neighborhood information aggregation to “compress” the
volume of information being transmit from theclusterheads
to the base station, furtherdecreasingenergydissipation and
improving system lifetime.
The procedure of LEACH is fragmented up into rounds,
where every round begins with a set-up phase, when the
clusters are structured, followed by a steady-state phase,
when data transfers to the base station occur. In order to
minimize overhead, the steady-state phase islongcompared
to the set-up phase.
3.1 Setup phase
In this phase clusters are framed and a cluster head (CH)
is decided for every cluster. Eachnodeproducesanarbitrary
number in the range of 0 and 1, and if that arbitrary number
is not as much as threshold value T (n), then it will become a
cluster head. In each round, T (n) is set to 0, for the sensing
element which previously functioned as CH in past rounds,
so that this node will not be chosen once more. The
possibility of being chosen is T (n) for the sensing elements
that have not been chosen once. If only single sensor node
left then T (n) is set to 1, implies this node will be surely
chosen as CH (Heinzelman, Chandrakasan, & Balakrishnan,
2002).
T (n) is characterized as below:
Where,
p= percentage of number of CH in the number of nodes in
network,
r = present round number,
G= set of sensing elements that have not been chosen in the
previous 1/p rounds of CH selection.
When any of the sensing elements is chosen as CH, it advises
different nodes. The nodes which are non-clusterheadspick
their CH in the view of received signal quality for this round.
The CH element sets up a TDMA schedule and transfers this
schedule to every one of the nodes in its cluster.

Fig.-2 Interval Period of LEACH
3.2 Steady-state phase
In this phase, the non-CH nodes begin detecting
information and send it to their CH as per the TDMA
schedule. The CH node aggregates the acknowledged
information and sends it to the sink. Communication is done
via direct-sequence spread spectrum and every cluster
utilizes a unique spreading code to decrease inter-cluster
interference. After certain timeframe, the system again goes
into the setup phase and enters another round of choosing
cluster heads (CHs).
3.3 Limitations of LEACH Protocol
A few of these assumptions are as per the following:
 All nodes can convey with sufficient power to reach
the base station if necessary.
 Nodes dependably have information to send.
 Nodes found near each other have connected
information. It is not evident how many
predetermined CHs are going to be uniformly
disseminated throughout the system. Therefore,
there is a possibility that the chosen CHs will be
focused in one part of the system. Subsequently,
some nodes will not have any cluster head closer to
them.
 CHs are chosen arbitrarily in LEACH, hence nodes
with less vitality might be picked up, and which
could lead to these nodes die too quick.
4.OVERVIEWOFPARTICLESWARMOPTIMIZATION
Particle Swarm Optimization (PSO) is encouraged by
natural lifecycles, like bird flocking, fish schooling and
arbitrary search techniques of evolutionary procedure
(Kennedy & Eberhart, 1995) (Wei & Nor, 2014). It can be
seen from the nature that creatures, especially winged
animals, fishes, etc. always travel together in a random
search for food in a cluster without colliding. It is because
every member tracks the cluster by modifying its position
and speed utilizing the cluster information.Thusitdecreases
individual's exertion for looking of nourishment, shelterand
so on.
PSO comprises of a swarm of a predefinedsize(sayNP)of
particles. Each element gives a complete answer to the
multidimensional optimizationissue.ThedimensionDof the
considerable number of particles is equal.AparticlePi,1≤i≤
NP has position Xid, 1 ≤ d ≤ D and velocity Vid in the dth
dimension of the hyperspace. The notation for
demonstrating the ith particlePioftheinhabitantsasfollows:
Pi = [Xi,1, Xi,2, Xi,3…….., Xi,D]
Every particle is estimated by a fitness function to judge
the superiority of the solution to the problem. Toachieve the
global best position, the particle Pi monitors its individual
best, i.e., personal best called Pbesti and global best called
Gbest to update its individual position and velocity. In every
repetition, its position Xid and velocity Vid and in the dth
dimension ismodifiedutilizingtheaccompanyingconditions.
Vi,d(t)= w * Vi.d(t-1) + c1r1 * (Xpbesti,d – Xi,d(t-1)) + c2r2
(Xgbestd – Xi,d(t-1))
Xi.d(t) = Xi,d(t-1 ) + Vi,d(t)
where
w = inertial weight
c1 and c2 = non-negative constants called acceleration factor
r1 and r2 = two different consistently disseminated arbitrary
numbers in the range [0,1].
The modified procedure is iteratively repeated until either
an acceptable Gbest is attained or a fixed number of
iterations tmax is achieved.
5. RESEARCH METHODOLOGY
5.1. Energy model
The radio model for energy which is used in this paper is
similar as discussed by W.B. et al. (Heinzelman,
Chandrakasan & Balakrishnan, 2002).Inthismodel,boththe
free space and multi-path blurring channels are utilized
relying upon the distance betweenthesenderandreceiver.If
the distance is not as much as threshold value d0, then the
free space (fs) model is applied, otherwise, the multipath
(mp) model is applied. Let Eelec, εfs and εmp be the energy
essential by the hardware circuit and by the amplifier infree
space and multipath respectively. Then the vitality required
by the radio to transmit an l-bit message over a distance d is
given as per the following:
ET(l,d) = lEelec+lεfsd2 for d < d0
lEelec+ lεmp d4 for d d0
The energy essential by the radio to accept an l-bit message
is given by
ER(l) = lEelec
The Eelec relies on few variables such as advanced coding,
modulation, filtering, and spreading of the signal, whereas
the amplifier energy, εfsd2/εmpd4, relies on the distance
between the transmitter and the receiver and also on the

acceptable bit-error rate. It ought to be observed that this is
an improved model. In general, radio wave propagation is
highly variable hard to demonstrate.
5.2. Network model and assumptions
In this paper, the proposed work simulates the WSN
model where all the sensing elements are deployed
arbitrarily along witha fewhighpoweredgatewaysandonce
they are deployed, theybecomestationaryforoneround and
vary from round to round. The process of Gateways
formation per round is same as the cluster head formation
process in LEACH. The assumptions made are following:
 All the chosen sensing elements are assumed as
static after deployment.
 Every one of the nodes can use power control for
various distances from the transmitter to the
receiver.
 Every one of the nodes is location unaware(i.e.they
are not armed with the GPS-gadgets).
 Sensor nodes are allocated with a unique
identification (ID) and similar preliminary power.
 The cluster-heads are dominant for performing
computations to the base station for long range
transmissions
 The base station utilizes the external energy supply
and the power will not be drained.
6. PROPOSED WORK
The proposed research work will perform the PSO with
leach protocol. Sensing elements are deployed arbitrarily
along with a few gateways. The proposed procedure is done
mainly in the following steps.
Step1: Apply PSO approach in clustering for optimal
selection of cluster head to improve the progression in the
remaining energy of node by sending a data packet to the
cluster head which is situated closesttotheBasestation. The
cluster head is nominated using PSO approach, based on the
distance from the cluster member node to base station and
the remaining energy of that node.
Step2: Compute the distance between transmitter and
receiver based upon the distance between the transmitter
and receiver. If the distance is less than a threshold value d0,
then the free space (fs) model is applied, otherwise, the
multipath (mp) model is applied.
Step3: A sensing element can be allocated to any gateway if
it is within the communicationrangeofthatsensing element.
Therefore, there are some pre-specified gateways on to
which an individual sensor node can be allocated.
The numerous steps of a PSO are represented in the
flowchart as shown in Fig. 3.
Fig.-3: Flowchart of PSO (Kuila & Jana, 2014)
The proposed algorithm:
Set-Up Phase
1. CH ══> N: idCH ,crc ,adv
2. ni ──> CH: idni , idCH ,crc, join_req
3. CH ══> N: idCH ,(… ,( idni , Tni)…) , crc ,sched
4. GW==> N: idgw, crc, adv
Steady State Phase
5. ni ──> CH : idCH , crc
6. ni -------> GW : idgw, crc
7. CH ──> BS: idCH , idBS , GW ----- > BS: idgw, idBS
8. Base Station get informationfromall theclusterheadsand
Gateways.
The symbol used in proposed algorithm signifies:
CH, ni, BS: Cluster Head, ordinary node, base station
N: Set of all nodes in the network
Adv,join_req,sched :String identifiers for message sorts
Crc : Cyclic redundancy check
idni , idCH , idBS :Nodes ni ,CH, BS id’s respectively
<y, Ty> : A node id y & its active slot Ty in the clusters TDMA
schedule
──>, ══>: Unicast, broadcast transmissions, respectively
7. EXPERIMENTAL SETUP
In this section, the simulation results for LEACH-P using
NS-2.34 are presented. WSN comprises of N = 100 and 200
sensors and few gateways which are arbitrarily deployed in
a field of measurement 100 m*100 m with a BS situated at
(50,175). Each sensing element was supposed to haveinitial
power of 2J and every gateway has 4.5J. The following
parameter values shown in Table1 in the simulation run. All
sensing elements are either static or micro-mobile are
considered and disregard the power lossduetocollisionand
intrusion between signals of dissimilar nodes.

The performance metrics used forthecomputationofthe
procedures are:stability period,systemlifetime,andnumber
of data transmitted to the BS.
 Stability period: By stability period, we mean how
much energy is dissipated in the network till the
entire network die or for how much time network
is stable.
 System lifetime: By system lifetime, we mean the
round number at which whole system die or the
number of rounds from network initialization till
the death of all nodes.
 Number of Data transmitted to BS: By this metric,
we mean the total number of data that are directly
transmitted to BS eitherfromCHsornon-CH nodes.
The parameters used in simulations are shown in
Table 1. Results along with discussions are givenin
the accompanying subsections.
Table-1: Simulation Parameters
Parameter Value
NS-2 Version 2.34
Channel Type Wireless Channel
Area 100*100
Routing Protocol LEACH
No of nodes 100 and 200
Number of cluster 5
Simulation Time 3600
Node initial energy 2 joules
Gateway initial energy 4.5 joules
Equal energy(startup) YES
RXThresh 6e-9
CSThresh 1e-9
Round Period Each 35 seconds
8. SIMULATION RESULTS
In this section simulationresultsforLEACHandLEACH-P
is depicted for 100 and 200 sensor nodes. Proposed
algorithm is simulated broadly and depicts the simulation
results for both the routing and clustering in a combined
way.
8.1 The Performance comparison of LEACH and
LEACH-P for 100 sensor nodes
Chart 1 shows the result analysis of total energy
dissipation in network for 100 sensor nodes. It is clear from
figure that LEACH-P consumes less energy than LEACH.
Chart 2 shows the result analysis of number of data
transmitted in LEACH-P is more as compared to LEACH as
gateways are used in LEACH-P to handle overload of dataon
sensor nodes. Chart 3 shows that the result analysis of
network lifetime. The first one node dies at 370 and 700
rounds for LEACH and LEACH-P respectively andapartfrom
first node, rest of nodes dies at 481 and 817 rounds for
LEACH and LEACH-P respectively. The simulation result
shows that LEACH-P lifetime is more than LEACH due to
cluster head optimization using PSO approach in LEACH-P.
Table-2: Performance comparison of LEACH and LEACH-P
for 100 sensor nodes
Performance
metrics
LEACH Proposed LEACH
Total Energy
Consumed(Joules)
309.101 433.04
Total Data
Transmitted(bits)
47863 85745
First Node Dies 370 700
Lifetime(rounds) 481 817
Chart-1: Result analysis of energy dissipation for 100
sensor nodes

Chart-2: Result analysis of no. of data transmitted for 100
sensor nodes
Chart-3: Result analysis of network Lifetime for 100
sensor nodes
8.2 The Performance comparison of LEACH and
LEACH-P for 200 sensor nodes.
Chart 4 shows how the energy dissipation in the system
varies as the percentage of sensors that are cluster-heads is
changed. From this plot, we find that the performance ofour
approach is superior over LEACH. LEACH-P consumes less
energy as compared to LEACH because of using PSO in
clustering for optimal selection of cluster head to enhance
the advancement in the residual energy of node bysendinga
data packet to the cluster head which is located very nearest
to the Base station. Chart 5 shows that the data transmitted
in LEACH-P is more as compared to LEACH. When the
routing and clustering is over, sensor nodes thoseare within
gateway communication send their data directly to gateway
and those are within cluster head communication sendtheir
data directly to their respective cluster head due to which
data transmitted is more than theLEACH.Chart6depictsthe
number of nodes alive and dead nodes in system lifetime.
The firs node for LEACH LEACH-P dies at 805 and 1310
rounds respectively and all nodes die at 1087 and 1858
respectively. It is clear from the figure 9 that LEACH-P is
superior to the LEACH.
Table-3: Performance comparison of LEACH and
proposed LEACH for 200 sensor nodes
Performance metrics LEACH Proposed
LEACH
Total Energy
Consumed(Joules)
866.1758 1260.719
Total Data
Transmitted(bits)
115398 194249
First Node Dies 805 1310
Lifetime(rounds) 1087 1857
Chart-4 Result analysis of energy dissipation for 100
sensor nodes
Chart-5 Result analysis of no. of data transmitted for 100
sensor nodes

Chart-6 Result analysis of network Lifetime for 100
sensor nodes
9. CONCLUSIONS
In this paper, the LEACH –P which is an enhancement of
LEACH is proposed. PSO in clustering is applied for optimal
selection of cluster head to improve the progression in the
residual vitality of node by transmitting a data packet to the
cluster head which is situated very nearest to the Base
station. Thus the power consumption of the CHs is
considerably balanced and the lifetime of the system is
enhanced. The experimental results have shown that the
proposed LEACH performs betterthantheexistingLEACHin
terms of system lifetime, stability period and the total data
transmission.
References
Abbasi, A., & Mohamad, Y. (2007). A survey on clustering
algorithms for wireless sensor networks. Comput.
Commun. 30, 2826-2841.
Abdul, L. (2007). Energy aware clustering for wireless
sensor networksusingparticleswarmoptimization.
IEEE PIMRC, 1-5.
Ahmad, A. (2012). MAC layer overview for wireless sensor
networks. CNCS 2012, (pp. 16-19).
Akkaya, K., & Younis, M. (2005, Novemeber). A Survey on
Routing Protocols for WirelessSensorNetworks.Ad
Hoc Networks, 3, 325-349.
Akyildiz, I. F., Su, W., Sankarasubramaniam, Y., & Cayirci, E.
(2002). Wireless sensor networks: a survey.
Computer Network Journal, 38(4), 393-422.
Al-Refai, H. (2011). Efficient routing LEACH (ER-LEACH)
enhanced on LEACH protocol in wireless sensor
networks. Int. J. Acad. Res., 42-48.
Aykut, Y. (2011). QoS-aware MAC protocols for wireless
sensor networks: a survey. In Comput. Netw (Vol.
55, pp. 1982-2004).
Bari, A., Wazed, S., Jaekal, A., & Bandyopadhyay, S. (2009). A
genetic algorithm based approach for energy
efficient routing in two-tiered sensor networks. In
Ad Hoc Networks (pp. 665-676). Elsevier.
Calhoun, B. (2005). Design considerations for ultra-low
energy wireless microsensor nodes. IEEE Trans.
Comput. 54, 727-740.
Gupta, G., & Younis, M. (2003). Load-balanced clustering of
wireless sensor networks. IEEE ICC 2003, 1848-
1852.
Habib, M., & Sajal, K. (2008). A trade-off between energyand
delay in data dissemination for wireless sensor
networks using transmission rangeslicing. Comput.
Commun.31, 1687-1704.
Heinzelman, W. B., Chandrakasan, A. P., & Balakrishnan, H.
(2002, October). An application-specific protocol
architecture for wireless microsensor networks.
IEEE Transactions on Wireless Communications,
1(4), 660-670.
Kennedy, J., & Eberhart, R. (1995). Particle Swarm
Optimization. IEEE International Conference on
Neural Networks, (pp. 1942-1948).
Kuila, P., & Jana, P. (2012). An energy balanced distributed
clustering and routingalgorithmfor wirelesssensor
networks. PDGC 2012, IEEE Xplore, 220-225.
Kuila, P., & Jana, P. (2012). Improved load balanced
clustering algorithm for wireless sensor networks.
ADCONS 2011, LNCS 7135, (pp. 399-404).
Kuila, P., & Jana, P. K. (2014, May). Energyefficientclustering
and routing algorithms for wireless sensor
networks: Particle swarm optimization approach.
Engineering Applications of Artificial Intelligence,
33, 127-140.
Kuila, P., Gupta, S., & Jana, P. (2013). A novel evolutionary
approach for load balanced clustering problem for
wireless sensor networks. Swarm Evol.Comput.12,
48-56.
Kulkarni, R. (2011). Particle swarmoptimizationinwireless-
sensor networks: a brief survey. IEEE Trans. Syst.,
Man, Cybern, 262–267.
Low, C. (2008). Efficient load-balancedclusteringalgorithms
for wireless sensornetworks. Comput.Commun.31,
750-759.
Saleem, M. (2011). Swarm intelligence based routing
protocol for wireless sensor networks: survey and

future directions. In Inf. Sci. (Vol. 181, pp. 4597-
4624).
singh, A. (2016). LEACH based energy efficient routing
protocol forwirelesssensornetworks.International
Conference on Electrical, Electronics, and
Optimization Techniques(ICEEOT), 4654-4658.
Singh, B., & Lobiyal, D. K. (2012). Energy-awareClusterHead
Selection Using Particle Swarm Optimization and
Analysis of Packet Retransmissions in WSN.
Procedia technology, 4, 171-176.
Tyagi, S., & Kumar, N. (2013). A systematic review on
clustering and routing techniques based upon
LEACH protocol for wireless sensor networks. In
Netw. Comput.Appl. (pp. 623–645).
Wei, H., & Nor, A. (2014). Particle swarm optimization with
increasing topology connectivity. In . Eng. Appl.
Artif. Intell. (Vol. 27, pp. 80–102).
Zungeru, A. (2012). Classical and swarm intelligence based
routing protocols for wireless sensor networks: a
survey and comparison. In Netw.Comput.Appl.(pp.
1508-1536).

Energy Efficient Clustering Protocol for Wireless Sensor Networks using Particle Swarm Optimization Approach

Recommended

More Related Content

What's hot (14)

Similar to Energy Efficient Clustering Protocol for Wireless Sensor Networks using Particle Swarm Optimization Approach (20)

More from IRJET Journal (20)

Recently uploaded (20)

Energy Efficient Clustering Protocol for Wireless Sensor Networks using Particle Swarm Optimization Approach