Architectural analysis for lifetime maximization and energy efficiency in hybridized WSN model

Abstract

It is well known that WSN is one of the leading techniques in granting pervasive computing for various applications regarding health sector and communication sector. However, the raising of issues in WSN is still a burden cause because of certain renowned terms like energy consumption and network lifetime extension. Clustering is a major contribution in any network and moreover Cluster Head selection is also a vital role since it is additively responsible in sending data to the base station, which means that Cluster Head directly makes its communication with base station. Day by day, the researches in cluster head selection get increased, but the requirements are not yet fulfilled. This paper proposes a energy efficient cluster head selection algorithm for maximizing the WSN lifetime. This paper develops a hybrid optimization process termed Group Search Ant Lion with Levy Flight (GAL-LF) for selecting the Cluster head in WSN. The proposed model is compared to the conventional models such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Group Search Optimization (GSO), Ant Lion Optimization (ALO) and Cuckoo Search (CS). The outcome of the simulation result shows the superiority of the proposed model by prolonging the lifetime of the network.

Full text

Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original work is properly cited.
International Journal of Engineering & Technology, 7 (2.7) (2018) 494-501
International Journal of Engineering & Technology
Website: www.sciencepubco.com/index.php/IJET
Research Paper
Architectural analysis for lifetime maximization and energy
efficiency in hybridized WSN model
Kale Navnath Dattatraya1*, K. Raghava Rao 2 , D.Satish Kumar3
1 K. L. University, Vijayawada, A.P., India
2 K. L. University, Vijayawada, A.P., India
3 Department of Mathematics, K. L. University, Vijayawada, A.P., India
* Corresponding author E-mail : piyushkale6@gmail.com
Abstract
It is well known that WSN is one of the leading techniques in granting pervasive computing for various applications regarding health
sector and communication sector. However, the raising of issues in WSN is still a burden cause because of certain renowned terms like
energy consumption and network lifetime extension. Clustering is a major contribution in any network and moreover Cluster Head selec-
tion is also a vital role since it is additively responsible in sending data to the base station, which means that Cluster Head directly makes
its communication with base station. Day by day, the researches in cluster head selection get increased, but the requirements are not yet
fulfilled. This paper proposes a energy efficient cluster head selection algorithm for maximizing the WSN lifetime. This paper develops a
hybrid optimization process termed Group Search Ant Lion with Levy Flight (GAL-LF) for selecting the Cluster head in WSN. The pro-
posed model is compared to the conventional models such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Artificial
Bee Colony (ABC), Group Search Optimization (GSO), Ant Lion Optimization (ALO) and Cuckoo Search (CS). The outcome of the
simulation result shows the superiority of the proposed model by prolonging the lifetime of the network.
Keywords: ALO; Cluster Head Selection; GAL-LF; Network lifetime; WSN
1. Introduction
Wireless Sensor Network (WSN) is an all-time hot research topic,
which steps its factual usage in many of the broad applications
that range from industrial and military applications, especially in
health sector and monitoring environmental strategies [3]. WSNs
include a huge count of sensor nodes (SNs) that are arbitrarily
deployed to sense and monitor both physical as well as environ-
mental conditions [4]. Current enhancements in WSNs lead to
various novel protocols that are particularly planned for sensor
networks where energy awareness is a necessary concern. Mostly
the attention, however, has been diverted for routing protocols as
they might get varied based on the various application as well as
network architecture.
Clustering is a method in which choosing of Cluster Head [7] [9]
[10] is done for preserving the energy consumption in WSN [4].
The network is separated into various groups namely clusters, and
this is the designing process of clustering technique. In each clus-
ter there presents a leader namely cluster head. In any clusters, the
head of the cluster is the only responsibility in collecting data
from their member sensors. Data aggregation is performed by
cluster head, and it eliminates the redundant data, which minimiz-
es the energy depletion of the network. Conversely, cluster heads
[6] [8] also consumes more energy due to extra overload for both
receiving as well as data aggregation and also to transmit data to
the sink [5]. Thus, the proper cluster head choosing plays a vital
role for energy conservation and enhancement of WSNs lifetime.
Moreover, cluster head selection is a optimization issue which is
the NP-hard problem. Number of algorithms is there in cluster
head selection, and some of them are Particle Swarm Optimization
(PSO), Low Energy Adaptive Cluster Hierarchy (LEACH) algo-
rithm and so on. Still, the problems raise in cluster head selection
are high energy consumption, security, increasing of link distance,
etc.
2. Literature Review
A. Related Works
In 2016, Gulnaz et al. [1] have developed a novel routing algo-
rithm termed ‘Sleep-awake Energy Efficient Distributed (SEED)
clustering algorithm.' There made a separation in the network
sensing field as the cluster head has direct contact with base sta-
tion. Cluster head in high energy region has greater distance
communication, and also it necessitates further energy cost when
compared to cluster head in the low energy area. With the consid-
eration of six criteria, they have examined the algorithm. Finally,
the investigations were done for testing the developed approach.
The investigational outcome has attained higher network lifetime
along with the greatest throughput when compared to other exist-
ing algorithms.
In 2016, Shankar et al. [2] have proposed a hybrid model of
‘Harmony Search Algorithm (HSA)’ as well as ‘Particle Swarm
Optimization (PSO)’ procedure for the efficient selection of clus-
ter head. The model has attained global search with rapid conver-
gence. With the developed hybrid model, the network lifetime has
greatly enhanced, and the performance evaluation was done with

International Journal of Engineering & Technology
495
the concern of some parameters like a number of alive nodes,
throughput, count of dead nodes, energy. The output of the result-
ant hybrid model has exposed an improvement in the energy and
the throughput by 83.89% and 29.00%, respectively, which was
better than the PSO algorithm.
Mahajan et al. [3] have stated that in any clustering based ap-
proach, one of the major concepts is the choosing of proper cluster
heads and also the development of balanced clusters as well. Ini-
tially, the cluster heads are chosen on the basis of weight metric,
and later the formation of cluster carries out. This model aims at
conserving sensor energy and load balancing. They have adopted
a local clustering approach within the cluster for minimizing the
evaluation rate and the cost of communication. Moreover, they
have explored a novel approach for transmitting data. The pro-
posed model has compared its performance with other conven-
tional techniques like LEACH, WCA as well as IWCA. The de-
veloped model has shown the greatest enhancement with respect
to lifetime as well as consumption of energy.
Khan et al. [4] have proposed a Fuzzy-TOPSIS approach, which
was on the basis of multiple criteria decision making for selecting
cluster head even more efficiently for maximizing the lifetime of
WSN. They have considered various criteria such as residual en-
ergy; consumption rate of energy node; neighbour node counts;
average distance among neighbouring nodes and distance from the
sink. In order to minimize the consumption of energy, they have
used a communication mechanism that was on the basis of intra-
cluster and inter-cluster. They have also examined the effect of
node density and various mobility strategy types for investigating
the impact on the lifetime of WSN. For maximizing the distribu-
tion of load in WSN, they have proposed a knowable mobility
along octagonal trajectory. The obtained results have enhanced the
overall network lifetime and latency.
O.Deepa and J.Suguna [5] have proposed a Optimized QoS-based
Clustering with Multipath Routing Protocol (OQoS-CMRP) for
WSNs that have minimized the consumption of energy in sink
coverage area via the application of Modified Particle Swarm
Optimization (PSO)-based clustering algorithm. This forms the
clusters for choosing the cluster heads in sink coverage area. They
have used the Single Sink-All Destination algorithm for finding
closer optimal multi-hop communication path. This was for the
neighborhood nodes. In order to transfer data to sink, they have
utilized Round-robin Paths Selection algorithm. As per QoS met-
rics, the proposed protocol has made its evaluation and also the
comparison was made to some of the conventional protocols like
EE-LEACH as well as EPSO-CEO. Finally, the output has shown
the superiority of proposed work over other models.
B. Review
The detailed review on energy efficient CH selection in WSN is
shown in with its respective features and challenges are given in
Table I. On the contrary, HSA and PSO-based CH selection [2]
attains faster convergence with improvement in residual energy
and throughput. However, it suffers from the requirement of more
number of iterations to get the optimum value, the problem of
poor convergence in refined search stage, and sometimes it gets
trapped in the local search area. In addition, [1] proposes the
SEED clustering algorithm with higher network lifetime and
throughput. However, the main issue of this technique is, it needs
to select new random seeds every time to minimize the risk. Thus
it is motivated to implement the energy efficient CH selection
based on the effective optimization algorithm. CCWM [3] is
applicable in number of sectors however, there requires more
consumption of energy if there presents more clusters. Fuzzy-
TOPSIS technique [4] is more robust in enhancing the network
lifetime, still, the enhancement is needed in the dealing of vague-
ness in the developed model. Moreover, OQoS-CMRP [5] could
attain better communication reliability; however, improvement is
needed in the aspect of ‘Security’. Thus, there require additional
protocols to solve all the above mentioned drawbacks.
Table 1: Review on energy efficient CH selection in WSN
Author [Citation]
Adopted Methodology
Features
Challenges
Gulnaz et al. [1]
SEED clustering algo-
rithm
❖ Higher network lifetime
❖ High throughput
❖ Require to select new random seeds every time
to minimize the risk
Shankar et al. [2]
HSA and PSO
❖ Faster convergence
❖ Improvement in residual energy
and throughput
❖ Needs to maximize the iteration to get the
optimal value
❖ Slow convergence in refined search stage
❖ Get trapped in local search area
Mahajan et al. [3]
Cluster Chain Weight
Metrics
approach (CCWM)
❖ Applicable in various sector
❖ Enhances the energy efficiency
❖ More clusters result in more energy consump-
tion
❖ Increase of link distance make worst perfor-
mance
Khan et al. [4]
Fuzzy-TOPSIS tech-
nique
❖ More robust
❖ Enhances the network lifetime
❖ Not so Flexible
❖ Improvement is needed in dealing with vague-
ness in the Fuzzy TOPSIS
O.Deepa and J.Suguna [5]
OQoS-CMRP
❖ Attains better communication
reliability
❖ Can maintain reasonable energy
consumption
❖ Multimedia messaging application is impossi-
ble
❖ Enhancement is needed in the security of pro-
posed model
3. Selection of cluster head in wireless sensor
network
Let’s take a WSN consideration that has
CS
N
number of
clusters, where the cluster is termed as
i
CS
,
where
CS
Ni ,.......1
. Here,
ij
NO
represents the node in
cluster, where
Ai ,.......2,1
and
Bj ,........2,1
. The cluster
head,
i
CH
is chosen from the presenting nodes in any cluster.
Moreover,
i
CH
leads the other nodes in cluster. While select-
ing the cluster head, there needs the consideration of certain
parameters like energy, packet delay and distance between
nodes present in cluster. Only the cluster head has direct con
tact with base station
s
B
. Furthermore, the choosing of cluster
head is not a simple task as it must consider some distinct
WSN characteristics. Hence, it is more significant to propose
an effective algorithm with parameter consideration, which
absolutely enhances the performance of network
Objective model
WSN parameters like distance and energy are used for doing
the selection of cluster head. Additionally, the proposed work
considers QoS as the vital parameter for the effective perfor-
mance of network. The network performance increases if there
have high energy and QoS, and low distance. The objective
model is determined in Eq. (1) and (2), where

denotes the
constant value, 0.3. In Eq. (2),
1

2

and
3

are specifying the

496
International Journal of Engineering & Technology
parameters of distance, energy and delay, respectively and
|||| sr BNO 
in Eq. (3) denotes the distance between the nor-
mal node and base station.
 
10;1 12
1

ddD
(1)
delay
i
energy
i
dist
idddd  321
1

(2)
||||
1
1
2sr
N
r
CS
BNO
N
dCS  

(3)
Internet
Sensor node
Cluster head
Base station
Fig. 1. Schematic diagram of cluster head selection in WSN
4. Optimal cluster head selection using hy-
brid optimization
A. Conventional GSO
In GSO [14], the population is denoted as ‘group,’ and the
individuals in population are termed as ‘member.' The visual
scanning field is generalized to
di
dimensional space and that is
considered by maximum
1max DI
and minimum pursuit
angle
1max 

. At
th
t
iteration,
PR
X
producer acts as speci-
fied below:
(i) Firstly,
PR
X
makes the scanning at zero degree and then, it
laterally scans in scanning field via sampling three points (arbi-
trarily): point at zero degree
ZE
X
is determined in Eq. (4), Eq.
(5) determines the point at right side hypercube
RI
X
and Eq.
(6) gives the point at left side hypercube
LE
X
, where
1
1n
is
the random number with mean 0 and Standard Deviation (SD)
1,
1
2
 r
n
is the dispersed random sequence in range (0,1),

denotes the head angle and
UV
signifies the unit vector.
 
tt
PR
t
PR
ZE UVDInXX  max
1
(4)
 
2/
max
2
max
1

nUVDInXX tt
PR
t
PR
RI 
(5)
 
2/
max
2
max
1

nUVDInXX tt
PR
t
PR
LE 
(6)
(ii)Then the producer finds the best point with best resource
(fitness). Moreover, if the best point has best fitness value than
the current position, then the producer makes a fly towards that
point. If the present position is superior than the fitness, then
the producer turns its head to a new randomly selected angle,
which is shown in Eq. (7), where
1max 

denotes the max-
imum turning angle.
max
2
1

n
tt 

(7)
(iii) even though with the completion of
i
iterations, and if the
producer cannot find the best location, then it turns its head
towards zero degree, which is determined in Eq. (8),
where
1
i
is a constant.
lil



(8)
B. Conventional ALO
ALO [15] aids the ‘hunting behavior’ of antlions. The algo-
rithm includes five major steps that are as follows: Ant’s arbi-
trary walk, Building the trap, Ant’s Entrapment, Prey catching,
rebuilding the traps. Moreover, the main phases of antlions life
cycle are ‘larvae’ and ‘adult.'
Arbitrary walks of ant: The Ant’s arbitrary walks are
persisted according to Eq. (9), (10) and (11), where
cusum
de-
notes the cumulative sum, iteration counts are determined
by
nu
,
 
sfu
denotes the stochastic function that is in Eq. (18),
and
st
indicates the arbitrary walk steps.
       
  








12........,
,...12,12,0 21
nu
stfucusum
stfucusumstfucusum
stAR
(9)
 





5.00
5.01
ranif
ranif
stfu
(10)
Moreover, the regularization of arbitrary walk is done for mak-
ing it within the search space, which is given in Eq. (11),
where
j
mi
and
j
ma
denotes the minimum and maximum of
arbitrary walk in
th
j
iteration and
s
j
k
and
s
j
l
indicates the min-
imum and maximum of
th
j
variable at
th
s
iteration respectively.
   
j
j
s
j
s
jjj
st
j
st
jq
mil
klmiAR
AR 



(11)
Antlion’s trap in Pit: The random walk of the ant is inclined
in the antlions’ trap, and the mathematical model is defined in
Eq. (12) and (13), where
s
k
specifies the minimum variable
at
th
j
iteration,
s
l
indicates the vector that includes maximum
variable at
th
j
iteration,
s
v
k
and
s
v
l
specifies the minimum and
maximum variables of whole variables for
th
j
ant,
and
s
v
antlion
refers to the position of chosen
v
antlion at
th
s
location.
ss
v
s
jkantlionv 
(12)

International Journal of Engineering & Technology
497
ss
v
s
jlantlionl 
(13)
Building the trap: Here, the hunting ability of antlion is
modeled with the utilization of roulette wheel. This is how the
ant lion gets picked under their fitness during optimization. The
respective model provides best potential to the fitting antlion to
catch the ants.
Descending ants towards antlion: The mathematical model of
this step is defined as given in Eq. (14), where
RA
indicates the
ratio, which is determined in Eq. (15), where
I
indicates the
progressing iteration, the maximum iterations is defined
as
max
I
, and
co
specifies the defined constant, which is based
on current iteration.
RA
l
l
RA
k
ks
s
s
s &
(14)
max
10 I
s
RA c

(15)
Prey catching and pit rebuilding: If the ant reaches in the
bottom of pit, then there comes an end for hunting, and it get
stuck in the jaw of the antlion. After the completion of this
phase, the antlion hit the trapped ant in the soil and slowly eats
its body. Then the location update of antlion happens with
newest one for enhancing the venture of grasping the new prey,
which is determined in Eq. (16), where
s
j
an
indicates the posi-
tion of
th
j
ant at
th
s
iteration.
   
s
v
s
j
s
j
s
vantlionFLanFLifanantlion 
(16)
Elitism: All ants should do a random walk around a distinct
antlion via roulette wheel as well as elite as given in Eq. (17),
where
s
AR
WA
indicates the arbitrary walk around the selected
antlion through the roulette wheel at
th
s
iteration,
s
EL
WA
speci-
fies the random walk around elite at
th
s
iteration.
2
s
EL
s
AR
s
jWAWA
an 

(17)
C. Proposed hybrid model
When considering of existing GSO, it is observed that it suffers
from certain issues like slow convergence along poor ability of
exploration and exploitation. Moreover, the existing ALO algo-
rithm also suffers from certain problems like greatest conver-
gence time for uncertainty, changes in probability allotment,
etc. To rectify all issues, this paper aims to develop a hybrid
model that is explained as follows: the input to this algorithm is
random solutions. The solution vec-
tor
 
CS
NEV CSCSCSSO ......, 21

implies a count of clusters.
Each cluster
 
ni NUNOCS ,.......
1

includes various nodes.
Here, two operations are processed parallel with the solutions
for choosing the proper cluster head. Best solution from the
random solutions (i.e., ‘cluster head’) gets updated as per Eq.
(4), (5), (6), (7) and (8), and the residual solutions are updated
as per Eq. (17). The model outputs the best solution as
best
HI
.
In the meantime, the input solutions are arbitrarily updated via
Levy model, and the corresponding solution is termed
as
best
LE
. Algorithm 1 gives the pseudo code of developed
hybrid model.
Levy Flight [33]: The levy fight characteristics are reviewed
from the flight behavior of various animals and insects. The
levy flight characteristics show its ability by means of optimal
search.
The average of both the
best
HI
and
best
LE
gives the optimal
cluster head
best
CH
, which is determined in Eq. (18)
2
bestbest
best LEHI
CH 

(18)
ALGORITHM 1: GAL-LF algorithm for Cluster head Selection
Best solution Updating via GSO
Set
0:t
Population size initialization the
sz
PO
, position initializa-
tion
i
X
and head angle initialization,
i

Fitness evaluation of initial members
While(if the stop criteria not satisfied)
For(each member
i
in group)
Choose the producer:
PR
X
Do Producing:
Evaluate the fitness of the present member
End for
Set
1 tt
End while
Residue solution updation using ALO
Population initialization of antlions and ants
Fitness evaluation of antlions and ants
The best antlions are positioned and supposed as the elite
While (stop criterion not satisfied)
For all ants
Antlion that uses roulette wheel is selected
update
k
and
l
using Eq. (14) and (15)
Production of random walk and normalization via Eq.
(9) and (11)
Ant position update the by Eq. (17)
End of
Fitness value evaluation
Antlion is replaced with its succeeding, and if antlion
becomes fitter with Eq. (16)
Elite update, if antlion outfit more apt than the elite
End while
Return elite
Returns the best solution
best
HI
Random solutions update via Levy Flight
Return the best solution
best
LE
Identify the Optimal Cluster head
best
CH
using Eq. (18)
5. Evaluation of parameters
The mathematical model of three parameters is determined
below:
Energy: The used energy in WSN is specified in Eq. (19),
where
)( i
NOEN
and
)( j
CHEN
specifies the energy of
th
i
normal node and energy of
th
j
cluster head, respectively.
)(
)(
ad
bd
denergy
energy
energy 
(19)

498
International Journal of Engineering & Technology
 
juENbd B
j
energy 

1
)(
(20)
 
BjCHENNOENjuEN A
ji
iji  

1;)()(1)( 1
(21)
 
 
)()()( 11 j
B
j
i
A
i
energy CHENMaxNOENMaxBud  
(22)
Distance: Eq. (23) gives the mathematical model of distance,
where
 
bddist
denotes the distance between normal node and
cluster head and between the cluster head and base station of
the network, which is defined as shown in Eq. (24) and
 
ad dist
specifies the distance between two normal nodes,
which is shown in Eq. (25). The value of
 
bddist
should be in
[0,1] range.
 
 
ad
bd
ddist
dist
dist 
(23)
 
sj
A
i
B
jji
dist BCHCHNObd   
 1 1
(24)

 
 A
i
B
jji
dist NONOad 1 1
)(
(25)
Delay: Node’s data transmission delay is determined in Eq.
(26), where the delay value in Eq. (26) should be among the
range [0, 1]. If the count of nodes in a cluster minimizes, then
the minimization of delay occurs. In Eq. (26), WSN cluster
head is determined by numerator value, and the total number of
nodes is specified by the denominator value.
 
A
CHMax
d
B
jj
delay 1

(26)
6. Simulation results
A. Simulation setup
The proposed GAL-LF of cluster head selection in WSN is
implemented in MATLAB 2015a. The simulation procedure of
the proposed CH selection comprise parameters such as
distance, delay, and energy of WSN devices. Here the nodes
were distributed within the WSN network of the area
mm 100100 
with base station at center. The initial ener-
gy
In
E
of the network has value 0.5, the energy of free space
model
fr
E
has
2
//10 mbitpJ
. Further, the energy of power
amplifier
power
E
is set as
2
//0013.0 mbitpJ
, the transmitter
energy
tr
E
is set as
2
//50 mbitnJ
, and the data aggregation
energy
Da
E
was set as
signalbitnJ //5
. The proposed CH
selection model implies Sensors that varied from 100, 125, and
150 and the clusters that varied from 10, 15 and 20, respective-
ly. The analysis was carried out for 2000 rounds by comparing
it with other conventional GA [11], PSO [12], ABC [13], GSO
[14], ALO [15] and CS [16] based cluster head selection mod-
els.
B. Alive nodes
In this paper proposed model is compared to other conventional
methods by analyzing the number of alive nodes over number
of rounds, which is illustrated in Fig 2, Fig 3 and Fig 4. Here,
the analysis is done by varying the sensors present and clusters
in network, which is already described in simulation setup. Fig
2 shows the graphical representation of alive node analysis
with 100 sensor nodes and all varied clusters. Fig 3 is the re-
sultant graph of alive node analysis for 125 sensor nodes and
all cluster variation [17-27]. Similarly, Fig 4 shows the graph-
ical illustration of alive node analysis for 150 sensor nodes and
all cluster variations. All the Figures (Fig 2, 3 and 4) show the
analysis in terms of number of nodes alive versus number of
rounds. As the number of rounds increases the number of alive
nodes gets minimized. It is clearly evident from Fig 2 (a),
initially the alive node count is 100 and as the number of
rounds increases the alive node count decreases and however,
the proposed GAL-LF at 2000 round holds 30 alive nodes,
whereas the remaining methods suffers from poor performance
with less alive nodes (10 for ABC, 12 alive nodes for GA, 17
alive nodes for PSO and so on). Similar performance is
observed from the remaining Fig 3 and 4.
C. Normalized energy
This section explains the performance analysis in terms of
normalized energy by comparing the proposed work over
conventional methods. Fig 5 reviews the graphical representa-
tion of normalized energy analysis with 100 sensor nodes and
all clusters. The resultant graph in Fig 6 is for the normalized
energy analysis for 125 sensor nodes and all cluster variation.
Similarly, Fig 7 reviews the graphical illustration of normalized
energy analysis for 150 sensor nodes and all cluster variations.
Here, From Fig 5, (a), it is proven that as the number of rounds
increases, the energy level decreases, but the proposed method
shows its better performance over other methods in terms of
higher energy level even in the final round 2000. The same
results are observed from all the graphs shown in Fig 6 and 7.

International Journal of Engineering & Technology
499
Fig. 2. Alive node analysis by comparing proposed model with conventional methods (a)Sensor node 100 and cluster 10 (b) Sensor node 100 and cluster 15 (c)
Sensor node 100 and cluster 20
(a)
(b)
(c)
Fig. 3. Alive node analysis by comparing proposed model with conventional methods (a)Sensor node 125and cluster 10 (b) Sensor node 125 and cluster 15 (c)
Sensor node 125 and cluster 20
(a)
(b)
(c)
Fig. 4. Alive node analysis by comparing proposed model with conventional methods (a)Sensor node 150 and cluster 10 (b) Sensor node 150 and cluster 15 (c)
Sensor node 150 and cluster 20
(a)
(b)
(c)

500
International Journal of Engineering & Technology
(a)
(b)
(c)
Fig. 5. Normalized energy analysis by comparing proposed model with conventional methods (a)Sensor node 100 and cluster 10 (b) Sensor node 100 and cluster
15 (c) Sensor node 100 and cluster 20
(a)
(b)
(c)
Fig. 6. Normalized energy analysis by comparing proposed model with conventional methods (a)Sensor node 125and cluster 10 (b) Sensor node 125 and cluster
15 (c) Sensor node 125 and cluster 20
(a)
(b)
(c)
Fig. 7. Normalized energy analysis by comparing proposed model with conventional methods (a)Sensor node 150 and cluster 10 (b) Sensor node 150 and cluster
15 (c) Sensor node 150 and cluster 20
7. Conclusion
This paper has proposed a GAL-LF algorithm for evolving energy
aware cluster head selection in WSN network. The simulation of
the proposed work was done with the consideration of energy,
distance, and delay of the sensor nodes in WSN network. After the
simulation work, the performance analysis took place by consider-
ing normalized energy and alive nodes of the selected cluster head.
The performance of the developed model was compared with con-
ventional GA, PSO, ABC, GSO, ALO and CS-based cluster head
selection models, and has proven the superiority.
References
[1] Gulnaz Ahmed, Jianhua Zou, Mian Muhammad Sadiq Fareeda
and MuhammadZeeshan, " Sleep-awake energy efficient
distributed clustering algorithm for wireless sensor networks",
Computers & Electrical Engineering, vol. 56, pp. 385-398,
2016.
[2] T.Shankar, S.Shanmugavel and A.Rajesh, " Hybrid HSA and
PSO algorithm for energy efficient cluster head selection in

International Journal of Engineering & Technology
501
wireless sensor networks", Swarm and Evolutionary
Computation, vol. 30, pp. 1-10, 2016.
[3] Shilpa Mahajan, Jyoteesh Malhotra and Sandeep Sharma, " An
energy balanced QoS based cluster head selection strategy for
WSN", Egyptian Informatics Journal, vol. 15, pp. 189–199,
2014.
[4] Bilal Muhammad Khana, Rabia Bilalb and Rupert Young, "
Fuzzy-TOPSIS based Cluster Head selection in mobile
wirelesssensor networks", Journal of Electrical Systems and
Information Technology, available online, 2016.
[5] O.Deepa and J.Suguna, " An optimized QoS-based clustering
with multipath routing protocol for Wireless Sensor Networks",
Journal of King Saud University - Computer and Information
Sciences, available online, 2017.
[6] ZhoujingYe, XinlongTong, HailuYang, LingjianMeng,
WenjingXue and LinbingWang, " The Strip Clustering Scheme
for data collection in large-scale Wireless Sensing Network of the
road", International Journal of Pavement Research and
Technology, vol. 11, no. 2, pp. 138-145, 2018.
[7] G.Kannan and T.Sree Renga Raja, " Energy efficient distributed
cluster head scheduling scheme for two tiered wireless sensor
network", Egyptian Informatics Journalm, vol. 16, no. 2, pp.
167-174, July 2015.
[8] Nandakishor Sirdeshpande and Vishwanath Udupi, " Fractional
lion optimization for cluster head-based routing protocol in
wireless sensor network", Journal of the Franklin Institute,
vol.354, no. 11, pp. 4457-4480, July 2017.
[9] AbdulhamidZahedi, " An efficient clustering method using
weighting coefficients in homogeneous wireless sensor
networks", Alexandria Engineering Journal, Available online,
2017.
[10] Alfonso de Hoyos, JavierPortillo, PilarMarín, F.Maestú,
LafuenteM and AntonioHernando, "Clustering strategies for
optimal trial selection in multisensor environments. An
eigenvector based approach", Journal of Neuroscience Methods,
vol. 222, pp. 1-14, 2014.
[11] P. Nayak and B. Vathasavai, "Genetic algorithm based clustering
approach for wireless sensor network to optimize routing
techniques," 2017 7th International Conference on Cloud
Computing, Data Science & Engineering - Confluence, Noida,
pp. 373-380, 2017.
[12] Q. Ni, Q. Pan, H. Du, C. Cao and Y. Zhai, "A Novel Cluster
Head Selection Algorithm Based on Fuzzy Clustering and
Particle Swarm Optimization," in IEEE/ACM Transactions on
Computational Biology and Bioinformatics, vol. 14, no. 1, pp.
76-84, Jan.-Feb. 1 2017.
[13] D.Karaboga and B.Basturk," On the performance of artificial bee
colony (ABC) algorithm", Applied Soft Computing, vol. 8, no. 1,
pp. 687-697, 2008.
[14] S. He,Q. H. Wu and J. R. Saunders, "Group Search Optimizer:
An Optimization Algorithm Inspired by Animal Searching
Behavior", IEEE TRANSACTIONS ON EVOLUTIONARY
COMPUTATION, vol. 13, no.5, 2009.
[15] N. Chopra and S. Mehta, "Multi-objective optimum generation
scheduling using Ant Lion Optimization," 2015 Annual IEEE
India Conference (INDICON), New Delhi, pp. 1-6, 2015.
[16] Xiin-She Yang and Deb, "Engineering Optimization By
CuckooSearch," J. Mathematical Modelling and Numerical
Optimisation,vol. 1, no. 4, 2010.
[17] SARMA, G.S., ASADI, S.S. and NARAYANA, S.L., 2016.
Creation of web based decision Support information system for
evaluation of Topographic characteristics using remote sensing &
GIS and visual basic programe. International Journal of Civil
Engineering and Technology, 7(6), pp. 621-634.
[18] MADHAV, B.T.P., KRISHNAM NAIDU YEDLA, G.S.,
KUMAR, K.V.V., RAHUL, R. and SRIKANTH, V., 2014.
Fractal aperture EBG ground structured dual band planar slot
antenna. International Journal of Applied Engineering
Research, 9(5), pp. 515-524.
[19] MADHAV, B.T.P., MOHAN REDDY, S.S., SHARMA, N.,
RAVINDRANATH CHOWDARY, J., PAVITHRA, B.R.,
KISHORE, K.N.V.S., SRIRAM, G. and SACHIN KUMAR, B.,
2013. Performance characterization of radial stub microstrip
bow-tie antenna. International Journal of Engineering and
Technology, 5(2), pp. 760-764.
[20] LAKSHMI, M.L.S.N.S., KHAN, H. and MADHAV, B.T.P.,
2015. Novel sequential rotated 2×2 array notched circular patch
antenna. Journal of Engineering Science and Technology
Review, 8(4), pp. 73-77.
[21] KISHORE, P.V.V., PRASAD, M.V.D., PRASAD, C.R. and
RAHUL, R., 2015. 4-Camera model for sign language
recognition using elliptical fourier descriptors and
ANN, International Conference on Signal Processing and
Communication Engineering Systems - Proceedings of SPACES
2015, in Association with IEEE 2015, pp. 34-38.
[22] MADHAV, B.T.P., PISIPATI, V.G.K.M., MADHAVI LATHA,
D. and DATTAPRASAD, P.V., 2012. Planar dipole antenna on
liquid crystal polymer substrate at 2.4 GHz.
[23] MADHAV, B.T.P., KAZA, H., VAKA, J.K., SRAVAN
KUMAR, K., SRIHARSHA, N., JASWANTH KUMAR, J.,
SIDDHARTH, D.S. and SAI TEJA REDDY, D., 2015. Design
and analysis of compact coplanar wave guide fed asymmetric
monopole antennas. Research Journal of Applied Sciences,
Engineering and Technology, 10(3), pp. 247-252.
[24] MADHAV, B.T.P., KHAN, H. and KOTAMRAJU, S.K., 2016.
Circularly polarized slotted aperture antenna with coplanar
waveguide fed for broadband applications. Journal of
Engineering Science and Technology, 11(2), pp. 267-277.
[25] MADHAV, B.T.P., PISIPATI, V.G.K.M., KHAN, H. and
UJWALA, D., 2014. Fractal shaped Sierpinski on EBG
structured ground plane. Leonardo Electronic Journal of Practices
and Technologies, 13(25), pp. 26-35.
[26] Sastry, J.K.R., Naga Sai Tejasvi, T., Aparna, J., Dynamic
scheduling of message flow within a distributed embedded
system connected through a RS485 network, ARPN Journal of
Engineering and Applied Sciences, Volume 12, Issue 9, 1 May
2017, Pages 2809-2817
[27] Sastry, J.K.R., Suresh, A., Bhanu, S.J., Building heterogeneous
distributed embedded systems through rs485 communication
protocol, ARPN Journal of Engineering and Applied Sciences,
2015, 10(16), pp. 6793-6803