IRJET- An Efficient Dynamic Deputy Cluster Head Selection Method for Wireless Sensor Networks

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 12 | Dec 2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 187
AN EFFICIENT DYNAMIC DEPUTY CLUSTER HEAD SELECTION METHOD
FOR WIRELESS SENSOR NETWORKS
Lohit B. Dalal1
1Student, ECE Dept. Acharya Institute of technology Bangalore, Karnataka, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract:- In view of the wireless sensor network
clustering algorithm at home and abroad, an Efficient
Dynamic Deputy Cluster Head Selection [EDDCH] technique
is used to solve the problem of unreasonable cluster head
selection, it may helpful to increase network life time and
reduce energy consumption of cluster communication
process. For data transmission,dynamic deputyclusterhead
selection is required to createthe network.Efficientdynamic
deputy cluster head selection is performed in viewof energy
level of node. The node with higher energylevel canbetaken
as Cluster Head [CH] and the node with next energy level as
Deputy Cluster Head [DCH]. After the selection of Cluster
Head and Deputy Cluster Head nodes, the information of
each individual node from the cluster is collected by the
cluster head. Further, the deputy cluster head advances the
information to base station. Cluster head and deputy cluster
head are turned in view of energy levels. Also the proposed
work confirms that the data transmission is effective and
energy efficient with increase in life timeofthe network.The
simulation outputs show that the proposed work balances
the network nodes energy compared with Dynamic Cluster
Head Selection Method [DCHSM] algorithm and expandsthe
lifetime of the network with decrease in energy utilization
between the hubs of the cluster.
Key Words: Cluster Head [CH], Deputy Cluster Head
[CH], Energy consumption.
1. INTRODUCTION
Wireless sensor network comprises many number
of little, light weight remote sensor hubs,havingconstrained
and comparable communication, processing, and detecting
capacities, deployed in expensive numbers to screen the
nature or system by the estimation of physical parameters
for example temperature, pressure or relativehumidity,and
to supportively transmit their information through the
system of fundamental area. Sensors are composed of are
comprised of miniaturized scale electro mechanical
frameworks (MEMS), every hub of sensor network
comprises of three subsystems: Sensor Subsystem,
Processing Subsystem, andCommunicationSubsystem.Now
day’s sensors are two directional, engages control sensor
activities. The improvement of remote sensor system is
utilized in military applications like war field observation
and such frameworks are utilized as a piece of few present
day and client oriented applications. Instance, Mechanical
system is checking & control, machine and machine
wellbeing watching etc. The Wireless sensor organize has
some awesome points of interests, for example flexible
communication and arrangement, minimum power
utilization and minimum cost are most utilized in modern
agriculture and criminal hunting [1].
The resources of the sensors determines the living
span of the sensor hub is shorter, and the wholesystemcan't
take care of the checking demand if a few hubs can't work
efficiently [2]. The main objective is to reduce the energy
utilization of the system but when the separation between
sensors increases the energy utilization of the network is
more & also maintaining with single cluster head because of
data load hence, while designing the wireless sensor
networks the energy utilization of the entirenetwork canbe
completely decreased when the hubs are composed in the
form of groups and also the group heads and dynamic
deputy cluster head is elected within the clusters depending
on energy levels of the sensor hubs to maximizes the life
time of the network[3].
2. LITRATURE REVIEW
Although more research activities are carried out to
enhance the performance of wirelesssensornetworksbutstill
they have limitations. The survey is conducted on all the
existing protocols and techniques used in WSN. The existing
protocols and techniques are like low energy adaptive
clustering algorithmandalsodynamicclusteringheadelection
techniques [4]-[5]. TheLEACH(lowenergyadaptiveclustering
hierarchy) it is an exceptionally understood various leveled
directing conventions which encourages random change of
group heads to uniformly adjust the energy utilization and all
information processing is carried out within the individual
group. But in this algorithm group head election is irregular.
There is chance of more non cluster member node, collision,
and routing issue [6]. Group head communicates with base
station in single hop mode which does not take part into
reducing the energy utilization and also the additional
overheads occurs because of change in cluster head. To
improve the performance of WSN they are using technique
called energy efficientdynamicdeputyclusteringalgorithm.In
this work group head is chosen depending on energy
heterogeneity. Here the cluster head does the information
collection from thegroupmembers&sendsaggregateddata to
sink node due to load in cluster head the life time of network
and performance of the network still not satisfactory. we are
taking power Consumption and performance of the network

into consideration we are going propose an efficient dynamic
deputy clustering algorithm in this work we choose the group
head & deputy group head depending on its energy levels and
also the multi hop packet transferring to the base station so
that we can decrease the energy utilization and we can
enhance the performance of the system. Along with that Multi
hop communication will increases the communication &
enhances the performance [7].
3. PROBLEM STATEMENT
Unreasonable cluster head election and energy
utilization is one of the serious issues in WSNs. This creates
tasks for industries and educational sectors. Hence energy
consumption handling is one of the key points to expand the
network life time. The main goal is to decrease the energy
usage of the network but when the distance between sensors
increases the energy usage of the network is more & also
maintaining with single cluster head because of data load
hence, while designing the wireless sensor networks the
distance can be fixed and also the cluster heads and dynamic
deputy cluster head is elected within the clusters on the bases
of energy levels of the sensor hubs to enhance the life time of
the system.
4. PROPOSED WORK:
The research about energyconsumption,lifetimeof
network and coverage of the network is done, but the
average energy utilization and the life time of the network
system still not satisfactory. In request to expand theenergy
efficiency and life time of the system. An efficient dynamic
deputy cluster head election technique for wireless sensor
system is proposed.
In this work wireless sensor networks, sensor hubs
are regularly gathered into individual distinctive gathering
called as group/cluster. Each group consist of group head
(CH), Deputy Cluster Head [DCH] and its members for
communication. CH and DCH are elected based on average
energy levels. Here Cluster Head Collect data from sensor
nodes and Deputy Cluster head performs information
accumulation and advancesinformationtoBase stationnode
i.e. Sink node. Figure 1 shows a tree diagram of group head
& dynamic deputy group head selection process. Grouping
it’s utilized as a part of WSNs sinceitgivediminishedgeneral
energy utilization, expanded coverage area, packet delivery
ratio and system lifetime.
Figure: 1 Tree diagram of Cluster Head & Deputy Cluster
Head Selection.
4.1 METHODOLOGY:
Figure: 2 the Procedure for Cluster Head & Deputy Cluster
Head Selection.
Steps involved in process of deputy cluster head
selection given below:
1. Initialization of network: the sensors hubs are
deployed in the monitoring zone and the
destination node can collects the location and
energy information of all the hubs.

2. The monitoring zone is divided in two numbers of
groups by vornoi diagram & the perception
probabilistic model is proposed.
3. Energy of all the hubs is estimated and hub which is
having higher energy within the group members
selected as cluster head in group one.
4. The weight of all the nodes is calculated by using
distance and energy of nodes.
5. The node having highest weight in cluster neighbor
groups is elected as cluster heads
6. The next energy level in cluster one is selected as
Deputy group head and weight of neighbor group
cluster is calculated and having highest weight
elected as deputy cluster head
7. After the election of group head node and deputy
group head, group head collects thedata fromevery
member of the cluster and Deputy group head
forwards data to destination and finally, the group
head and Deputy group Headarerotateddepending
on weight & energy level of nodes.
Packet Delivery Ratio (PDR):
The packet delivery ratio is defined as the ratio of packets
received at base station and the number of packets
transmitted/generated at source. Packet delivery ratio
(PDR) equation (1) can be given below:
Received Packets
PDR= --------------------- (1)
Generated packets
Average Residual-Energy:
The total amount of energy remaining in a hubatthepresent
instance of time is calledasremaining energy.The remaining
energy of network calculated by using following formula (2)
given below:
∑(Ie-Ec)
Residual Energy = ------------------------------ (2)
Number of packets transmitted
Where Ec is the energy consumed by nodes, Ie is Initial
energy of hubs and is the Residual energy of hubs.
Figure: 3 Flow chart for cluster head selection
5. SIMULATIONOUTPUTS
 Packet delivery ratio:
The Packet Delivery Ratio (PDR) is defined as the ratio off
how much number of packets are transmitted to the
destination node from the packets generated at the source,
i.e. transmitted. In existing system we had a problem of
unreasonable cluster head selection and cluster head over
load that will avoid packets from being sent to base station
node. So that’s the reason Packet Delivery Ratio will
diminish but in proposed system we are properly selecting
group head and deputy group head depending on energy
levels that will avoid the unnecessary node failure to sink
node. So that we can increase the PDR which is shown in
figure 4.We have maximum PDR for proposed EDDCHSM
(shown as red color) than the existing ones.

Figure: 4 Packet Delivery Ratio.
Table 1: Packet Delivery ratio
Time(ms) 0 9 11 13
EDDCHSM 0 1.0 1.0 1.0
DCHSM 0 0.5 1.0 1.0
The above Table shows the comparison between the two
cluster head selection methods namely, Efficient Dynamic
Deputy Cluster Head Selection Method (EDDCHSM) and
Dynamic Cluster Head Selection Method (DCHSM) with
respect to time. By looking the values of PDR (Table 1) in
both methods, it can be said that the EDDCHSM is efficient
than the DCHSM.
 Average-Residual Energy
The amount of energy remaining in node with respect to
certain time interval is calledasresidual energy.Theaverage
remaining energy graph is shown in figure 5. In existing
systems there is chance of more non cluster member node,
collision, and routing issue. Group head communicates with
base station in single hop mode which does not take part
into reducing the energy utilization butinproposedwork we
are selecting cluster head and deputy clusterheadsbased on
energy levels and also using multi hop routing will reduces
the extra energy consumption while transmission to base
station.
Hence the average remaining energy ismorewhichisshown
in figure 5. We have more remaining energy for proposed
EDDCHSM (shown as red color) than the existing ones.
The average residual energy values of nodes after the
simulation are written in the table 2. By comparing both
energy values of cluster head selection methods, it is seen
that the EDDCHSM is better in consumptionoflessenergy.In
other words more residual energy can be saved.
Figure: 5 Average residual Energy of nodes
Table 2: Average residual Energy of nodes
Time(ms) 2 6 8 14
EDDCHSM 98.8462 94.860 93.64 86.53
DCHSM 98.840 93.86 92.64 85.53
 Throughput
Throughput is defined as successful packet delivery rate in
bits per seconds. The packet delivery rate graph can be
plotted with respect to time as shown in the figure 6. By the
selection of two cluster heads (CH & DCH) in the group of
nodes, the occurrence of the cluster head overload is maid
minimum based on energy levels. As per this concept of
cluster head selection method, the performance of
throughput is made high compared to Dynamic cluster head
selection method (DCHSM).

Figure: 6 Throughput
Table 3: Throughput
It is seen that the throughput i.e. the packet delivery Success
rate of proposed work (EDDCHSM) is high compared to
Dynamic cluster head selection method (DCHSM).the above
table 3 gives the throughput values of both methods to say
that EDDCHSM is better in performance wise.
 Packet drop
Packet Drop can be defined as the number of packet loss or
dropped with respect to time. The figure 7 shows the plot of
packet drop of the both cluster head selection methods with
respect to time. The packet drop orpacketlossisalmostzero
in proposed work (EDDCHSM) compared to the existing
system (DCHSM).
Table 4: Packet Drop
Time(ms) 0 5 7 9 11 13
EDDCHSM 0 0 0 0 0 0
DCHSM 0 0 0 0.001 0 0
The above table 4 shows the packet drop values of the both
cluster head selection methods. It is clearly seen that the
packet loss in EDDCHM is comparatively Zero with DCHSM.
So it can be said that the proposedwork (EDDCHSM)isgood.
Figure: 7 Packet Drop
6. CONCLUSION
An efficient dynamic deputy cluster head selection method
for wireless sensor networks is proposed in this work by
analyzing the sensor network energy consumptionbasedon
energy heterogeneity. Theproposedwork mainlyfocused on
four major aspects, including average residual energy of the
network, packet delivery ratio, throughput and packet drop.
The performance is compared with dynamic cluster head
selection method (DCHSM) algorithm, it is shown that, this
method overcomes the imbalance of energyusage,improves
information redundancy in the process of transmission and
increases the life time of the network.
.
REFERENCES
[1] C. Alippi, G. Boracchi, R. Camplani, and M. Roveri,
“Wireless sensor networks for
monitoring Vineyards” in Methodologies and
Technologies forNetworked Enterprisesvolumee7200.
Berlin, Germany Springer, 2012 pp.120-123.
[2] Dongyao Jia, Huaihua Zhu, Shengxiong Zou,and Po Hu
“Dynamic Cluster Head Selection Method for Wireless
Sensor Network” ieee sensors journal, volume. 16, NO.
pp 8, APRIL 15, 2016, pp.2746
[3] R. C. Shah and J. M. Rabaey, “Energy aware routing for
low energy ad hoc sensor networks,” in IEEE Wireless
Time(ms) 0 5 7 9 11 13
EDDCHM 0 0 0 0.7167 0.7167 0.7167
DCHSM 0 0 0 0.6167 0.7167 0.7167

Communication.Network Conference(WCNC),Orlando,
FL, USA, March. 2002, pp. 350–355.
[4] A. Lindgren, A. Doria, and O. Schelén, “Probabilistic
routing in intermittently connected networks,” Mobile
Computer Communication. Volumee7, no. 3, pp.19–20,
July. 2012.
[5] S. Isik, M. Y. Dumez, and C. Ersoy, “Cross layer load
balanced forwarding schemes for video sensor
networks,” Ad Hock Network, lvolumee9, no.3,pp.265–
284, May 2011..
[6] W.-Y. Chung, B. G. Lee, and C. S. Yang, “3D virtual viewer
on mobile device for wireless sensor network based
RSSI indoor tracking system,” Sens. ActuatorsB,Chem.,
vol. 140, no. 1, pp. 35–42, June. 2009.
[7] M. Mafuta by “Successful deployment to wireless sensor
network for precision agriculture in Malawi,” in Proc.
IEEE Int. Conference Network. Embedded System
Enterprise Application, June 2012.
[8] D. D. Geeta, N. Nalini, and R. C. Biradar, “Fault tolerancein
wireless sensor network using handoff and dynamic
power adjustment approach,” J. Network Computer.
Appl. vol. 36, no. 4, pp. 1174–1185, July. 2013.
[9] S. Poduri and G. Sukhatme, “Constrained coverage for
mobile sensor networks,” in Proc. IEEE Conference.
Robot Automation. (ICRA), April./May 2004, pp. 165–
171.
10] H. Pan, H. F. Zhang, and Y. Li, “Energy efficient target
tracking algorithms in wireless sensor networks,” J.
Computer. Res. Develop., volume. 47, no. 2, pp. 64–68,
August 2010.
[11] S. Pino-Povedano, R. Arroyo-Valles, and J. Cid-Sueiro,
“Selective forwarding for energy efficient target
tracking in sensor networks,” Signal Process.,volume.
94, pp. 557–569, January. 2014.
[12] H. Yang and Y. Zhang, “A study of super capacitor
charger distribution for applications in
environmentally powered wireless sensor nodes,” J.
Power Source. vol. 273, pp. 223–236, January 2015.
[13] L. M. Sun Wireless Sensor Network. Beijing, Chinak T
singhua Univ. pp.29 Press 2005
[14] V. Mhatre and C. Rosenberg, “Design guidelines for
wireless sensor networks:Communication,clustering
and aggregation,” Ad Hoc Network, volumeg2, no. 1,
pp. 45–63, January. 2004.

IRJET- An Efficient Dynamic Deputy Cluster Head Selection Method for Wireless Sensor Networks

More Related Content

What's hot(20)

Similar to IRJET- An Efficient Dynamic Deputy Cluster Head Selection Method for Wireless Sensor Networks(20)

More from IRJET Journal(20)

Recently uploaded(20)

IRJET- An Efficient Dynamic Deputy Cluster Head Selection Method for Wireless Sensor Networks