![www.ijcrset.com 13 |
PPage
International Journal of Current Research In Science, Engineering and Technology (IJCRSET)
Volume 1, Issue 4, August 2016, PP 13-17
www.ijcrset.com
A Survey on Energy Balanced ClusteringApproach for
Homogenous Sensor Network
Ms. Neha D. Desai1
, Prof. Shrihari D. Khatawkar2
1
(Department of Computer Science and Engineering, ADCET, Shivaji University, India)
2
(Department of Computer Science and Engineering, ADCET, Shivaji University, India)
Abstract: Energy consumption is one of the significant characteristic should be considered within designing
of routing protocol. A one of major challenge is to improve the lifespan of network within designing of the
homogenous sensor network. This paper deal with a new approach for homogenous sensor network. This
method intended to balance the consumption of energy at member node as well as head node of cluster during
the data transmission stage. The aim of the intended method to improve energy efficiency and lifespan of
network.
Keywords – Wireless Sensor network, Energy Consumption, Tree structure, Lifespan of network.
I. INTRODUCTION
A wireless sensor network (WSNs) are organized by different number of sensor nodes randomly set up
in the target area [1]. A wireless sensor network consist of the varied range of application such as military,
hospital, planetary investigation and farming [2]. The homogenous sensor networks are the sensor network in
which all of sensor node have equal hardware capability and initial battery capacity.
A WSN nodes having restricted resources such as limited energy and restricted processing capabilities
[3]. The designing of routing protocol involves many challenges. The sensor node contain two constraints that
are restricted within supply of energy and bandwidth which are accumulates to decide the sensor network design
and management on basis of scalability.
In most design of WSN, data transfer is performed from member node of cluster to the head node of
respective cluster and then forwards to the base station. The design requirement needs to set up effective data
transmission technique order to balance the consumption of energy at member node of the cluster and head node
of the cluster [4]. There are many routing protocols have been introduced in WSN through they are lagging
within their performance, Hence there is need to work out with new techniques for change in the routing
protocol.
A tree structure is non-linear structure that can be derived from real tree [5]. A tree structure is usually
viewed from the upper side to down with its parent at the top and children on the bottom. A tree is typically
accessed from its root as parent node, then through its branches as child nodes. Hence there is need to balance
the energy consumption due to the data transmission stage. For this purpose, tree structure is implemented in
proposed approach. Each member node of cluster sends their data to the nearest parent node until it reaches to
the base station. The next task is to gather the data from member node of cluster at head node in energy efficient
way in order to improve the lifespan of the network [7]. Data aggregation is process of merging of data arriving
from a verity of sources by removing redundant information and compress them which will helps in balancing
the consumption of energy.
This paper is intended a routing protocol that is energy balanced clustering approach, which deploying
tree structure for homogenous sensor network [8]. This will minimize the amount of energy spend at
transmission of data among different sensor node, head node and base station which leads to reduce overhead of
re-clustering and helps to improve the data transmission rate and lifespan of network.
II. RELATED WORK
Research work done by previous author is discussed as below
In paper [6], W.R. Heinzelman purpose a clustering based protocol that is Low-Energy Adaptive
Clustering Hierarchy (LEACH) protocol. This protocol develops randomized rotation of native cluster’s base
stations which simply called as head node of related cluster. The head node consistently issues the energy load
between the nodes within the sensor network. In this, head node will be directly send data packets to the base
station, which are gathered from member node of cluster. In terms of complex network this approach is not at all
times possible. This will cause overhead of re-clustering, more energy consumption at head node.
Within [9], the author Kyung Tae Kim chooses a best number of clusters by organizing unique
technique which sets up threshold rate that involves the possibility of the best number of head node as well as
nodes remaining energy. The proposed scheme introduces the different method to improve the lifetime of the](https://image.slidesharecdn.com/2-160906050745/85/2-a-survey-on-energy-balanced-clustering-approach-for-homogenous-sensor-network-1-320.jpg)
![www.ijcrset.com 14 |
PPage
International Journal of Current Research In Science, Engineering and Technology
network by the construction of tree in each cluster. In this approach, the base station is placed a distant away
from the target area and the cluster head sends data directly to the base station without considering how much
distance between base station and head node. In this, there is more energy spent on transmission of data packet
from the head node to base station and degrade the lifespan of network.
In paper [10], K. T. Kim and H. Y. Youn introduced an enhanced LEACH. For these purpose author
use a method in which selects a proxy node which play part of the existing head node of the respective cluster
with weak power within one of iteration of communication. For discovery and management of failure of head
node, this method depends on healthy nodes. In this scheme, the head node is elected randomly which will make
uniform distribution within the network. Therefore, energy spend on head node as well as the base station. In
this, there may be possibilities of more energy load on head node will quicklyexpire.
According to [11], the author Y. HU, W. Li, Z. Kang discovers the node as head node, which having of
more energy with higher possibilities as compared to other nodes. Whenever they interact with the base station
then head node are set to make the usage of multi-hop passing as well as routing. In case of this method, if the
base station is far away from the head node, then more time and energy spend during transmission of data
packet. This leads to degrade the lifespan of network as well as energy efficiency of head node.
III. PROPOSED SCHEME
1. Problem Statement
Within the homogenous sensor network, nodes which are randomly deployed in target area. However,
each of node wish to send their redundant data packet over numerous routes. If these data packets are redundant
that will cause an energy deficiency of nodes and the lifespan of the entire network. For this problem, nodes are
organized into a set of nodes as cluster. Picking head node of cluster randomly rises the problem in which
existing energy of certain head node is lower or having larger distance between the base station and head node
will quickly expire. For this purpose, head node is selected based on the highest remaining energy of a node, the
number of adjacent nodes of a node, minimum distance between the base station and node [12].
In most of the procedure, all of node sends their data directly to head node without considering how
much distance between the member node and head node which may lead to energy deficiency of head node. For
this purpose, tree structure is used in which member node sends their data packet to parent neighbor node until it
reach to head node and also location based cluster joining in which threshold sensing range is used to reduce the
size of cluster. In this way, energy efficiency of head node will improved.
Afterwards, head node removes redundant data from accepted data packet, compress them forwards it
nearest route node which is also closest to the base station. In order to improve energy efficiency, lifespan of the
network, the head node calculates weight rate of its adjacent node. The weight rate is based on the remaining
energy of adjacent node, number of adjacent node of the current adjacent node as well as the distance between a
node and the base station, the energy consumption of node, count of data packet already received by current
node from the adjacent node. In this way, the head node forwards their gathered data to the base station in an
energy efficient way within less time.
The new energy balanced clustering approach aims at balancing energy consumption of the entire
network as well as enhancing life span of network by balancing energy consumption at head node and the
member node of the cluster during a data transmission stage.
2. Proposed Approach
The intended approach which organize node of the network in the form of tree structure. Within the target
zone, this will helps to balance the energy capacity spent on the nodes of the network. The five stages are
considered within these proposed approach.
2.2 Deployment of Homogenous Sensor Network
In proposed system, sensor nodes are randomly deployed in the network, which holds random topology. All
sensor begin with same initial energy. All nodes homogenous and have the identical capabilities. Each node is
assigned a unique identifier. The base station is located at the center of the target area.
2.3 Formation of Cluster and Selection of cluster head
The homogenous sensor network will be divided into an ideal number of clusters. The design of
establishment of cluster is based on tree configuration. Within this stage, some of nodes are randomly selected
as originator node. Any of the originator node with greater remaining energy will be playing role of
HighEnergy_originator node.
Therefore, the originator node will send the association request message to adjacent nodes within two hop
distance. Afterwards, all node of homogenous network should accept the association request from originator
node in order to form a cluster. The node which belongs to cluster are called as a member node of the cluster.](https://image.slidesharecdn.com/2-160906050745/85/2-a-survey-on-energy-balanced-clustering-approach-for-homogenous-sensor-network-2-320.jpg)
![www.ijcrset.com 15 |
PPage
International Journal of Current Research In Science, Engineering and Technology
Within next stage, the selection of the head node is implemented. The election of head node lies on greater
energy, lowest distance from the base station, threshold rate and remaining energy of node in that cluster. The
member node of the cluster becomes an applicant node of head node in election procedure if the remaining
energy of node is larger than the threshold rate. The threshold rate can be defined as minimum energy required
for accepting data packets from all nodes of the network, gathering them and forwarding that to the adjacent
node.
Each of applicant node wants to become a head node itself. For these purpose, there is need to calculate
these cluster values for each node of the network as
C
However,
C is node m’s cluster value.
is node m’s remaining energy.
are node m’s neighbor node count.
After this calculation, every member node sends its individual cluster value to its neighbor node. Then,
each member node of cluster accepts the cluster value of all neighbor nodes in order to distinguish that from its
individual cluster value to become a head node of related cluster. The node which contains, the greater cluster
value and lowest distance from base station will send an advertising message should send to its neighbor nodes.
After receipt of advertising message, its neighbor node decides their neighbor node through
considering parameter as received signal strength of different head node. For the purpose of decreasing the
cluster size, the threshold sensing range is used. It is expected that the received value of TRs that is based on the
sensing area of nodes. Therefore all member nodes of the associated cluster have sent their association request
to head node that is closer to them and also send their remaining energy, their locations. After selection of head
node, all member nodes are informed about the head node of recent iteration
2.4 Aggregation of data at cluster head
Once the formation of the cluster and selection of cluster head is complete then the next stage is aggregation
of data at cluster head. The nodes which are present within the sensing range of the particular head node will
initiate sensing stage. Therefore, every member node of cluster sends the resultant sensed data to the respective
head node of cluster. Therefore, the head node of the particular cluster eliminated the redundant data, gather
data and compress the resultant data.
2.5 Data Transmission towards base station
The next stage is the data transmission towards base station. Therefore, the gathered data packet from the
head node of related cluster, which will be forwarded to the base station in an energy efficient way in less time.
For this purpose, the head node will compute a weight rate for all neighbor nodes which have a distance from
the base station lesser than itself to send the gathered data [13]. The weight rate of neighbor node m should be
computed by node n is specified as
However,
E is the neighbor node m’s remaining energy.
T1 is a constant value.
is the number of neighbor nodes of neighbor node n.
is the distance between base station and node n.
is the node n’s energy spent rate.
is node m previously received number of data from node n within the recent iteration.
The node will be directly send data packet to the base station if and only if these nodes are nearest
neighbors to the base station. For every member node of the cluster transmit their sensed data to its parent node
within the assigned period. This process continuous until data transmission of all nods is completed. In
proposed system, sensor nodes are randomly deployed in the network, which holds randomtopology.
2.6 Performance Analysis Parameter
The performance of the proposed scheme will be evaluated as below](https://image.slidesharecdn.com/2-160906050745/85/2-a-survey-on-energy-balanced-clustering-approach-for-homogenous-sensor-network-3-320.jpg)
![www.ijcrset.com 16 |
PPage
International Journal of Current Research In Science, Engineering and Technology
2.6.1 Consumption of energy
One of performance measure is the consumption of energy in which energy spend within different
procedures such as transferring data packet and electing head node within each iteration. Hence, total
consumption of energy can be evaluated as the sum of consumption of energy of the all nodes within the target
area.
2.6.2 Packet Delivery Ratio
This performance metric is association amongst the total count of data packet accepted by the base
station will be distinguished from total count of data packets sent by all nodes of the network.
2.6.3 Throughput
Throughput is one of performance metric in which base station accepting a number of data packets.
IV. PROPOSED ARCHITECTURE
Base Station
Cluster Head
Cluster Member
Routing
Route_node
Sensing Range
Fig 1: Formation of Cluster in Homogenous Sensor Network
A group of variety of numbers of sensor nodes which have equal hardware capability and initial energy
capacity is termed as homogenous sensor network [1]. These sensor nodes are interacting with each other as
well as base station. In order to reach more accuracy over a larger geographical area, an enormous number of
nodes permit sensing within the associated target area. Figure 1 shows a representation of diagram of formation
of cluster within the homogenous sensor network. In each of the cluster there is one of node elected as a head
node using maximum cluster and lowest distance from the base station. The remaining nodes, which are within
the sensing area act as member node of that cluster. Afterwards, each member node of cluster sends their sensed
data to parent node and so on until that data packet reaches their head node. After receiving sensed data from
member node, the head node aggregate data, eliminate the redundant data and compress them [7]. The head
node forwards that resultant data to its parent node with highest weight rate and so on until that data packet
reaches the base station. In this way, the base station gathers data form all of the head nodes within the target
area
V. CONCLUSION
A Therefore, the selection of head node randomly within a homogenous sensor network causes that the
existing energy of a certain head node is lower or their distance from base station are far away. The larger
energy load on the head node will quickly expire. For this problem, the new energy balanced clustering method
which aims at balancing energy spend on the entire network and prolong lifespan of homogenous network by
balancing consumption of energy at head node of respective cluster and member node of cluster during the data
transmission stage.](https://image.slidesharecdn.com/2-160906050745/85/2-a-survey-on-energy-balanced-clustering-approach-for-homogenous-sensor-network-4-320.jpg)
![www.ijcrset.com 17 |
PPage
International Journal of Current Research In Science, Engineering and Technology
REFERENCES
[1] Long Zhaohua, Li Jianfeng, Jiang Gujquan, Survey of routing protocols in wireless sensor networks based on application, 2nd
International conference on industrial and information system 978-1-4244-8217-71102010 IEEE.
[2] Riyazahamed A Jamadari, Mousami S. Vanjale, Threats and attacks analysis of routing protocols for wireless sensor networks,
International Journal of Science and Research (IJSR) Vol.3 Issue 12, December. 2014.
[3] Shekhar kumar, Pravinash, survey on energy efficient protocol for wireless sensor network, International on conference of computer
application (0975-8887), Vol. 120- No.02, June2015.
[4] S. Roy, Kumar Das A., energy efficient cluster basd routing protocol (EECBRP) for wireless sensor network, network and soft
computing (ICNSC), 2014 first International conference, Page No. 20-29 DOI:1,1109/CNSC.214.690660.
[5] M. Lee and V.M.S. Wong, energy aware spanning tree algorithm for data aggregation in wireless sensor networks, IEEE PacRrim
2005.
[6] W. R. Heinzelman, A. Chadrakasan and H. Balkrishna, energy-efficient communication protocol for wireless sensor network, In
Proc.HICSS, pp.1-10, Jan.2000.
[7] Ramesh Rajagpalan and Pramod k. Varshney, data aggregation technique in sensor networks: A Survey, IEEE Communications
Surveys & Tutorials 4th
Quarter 2006, Vol.8.
[8] Mrityunjay, Singh, NiranjanLal, Monika sethi and Saroj Poonia, a tree based routing protocol for mobile sensor networks (MSNs)”,
(IJCSE) International Journal on Computer Science and Engineering, Vol.02, No .01S, 2010, 55-60.
[9] Kyung Tae Kim, Man Youn Kim, Ji Hyeon Choi, Hee Yong Youn “An Energy Efficient and Optimal Randomized Clustering for
Wireless Sensor Networks” IEEE SPND 2015, June1-3-2015.
[10] K. T. Kim and H.Y Youn “PEACH: Proxy-Enable Adaptive Clustering Hierarchy for Wireless Sensor Network”, In proc. ICWN,
pp. 5257, June 2005.
[11] Y. Hu, W. Li, Z. Kang, “Study on Energy Efficient Hierarchical Routing Protocol of Wireless Sensor Network”, In proc. ICIE, pp.
325-328,2009.
[12] JaimeLortMauri, Sabu M., Thampi, Danda B. Rawat, Di Jain, “Security in Computing and Communications: Second International
Symposium”, Pg.20-29, Sept 2014.
[13] Sohoni Roy, Ayankumar Das, “Cluster based Event Driven Routing Protocol Wireless Sensor Network”, International Journal of
Computer Application (IJCA), Vol.88, Feb2014.](https://image.slidesharecdn.com/2-160906050745/85/2-a-survey-on-energy-balanced-clustering-approach-for-homogenous-sensor-network-5-320.jpg)
2.a survey on energy balanced clustering approach for homogenous sensor network
- 1. www.ijcrset.com 13 | PPage International Journal of Current Research In Science, Engineering and Technology (IJCRSET) Volume 1, Issue 4, August 2016, PP 13-17 www.ijcrset.com A Survey on Energy Balanced ClusteringApproach for Homogenous Sensor Network Ms. Neha D. Desai1 , Prof. Shrihari D. Khatawkar2 1 (Department of Computer Science and Engineering, ADCET, Shivaji University, India) 2 (Department of Computer Science and Engineering, ADCET, Shivaji University, India) Abstract: Energy consumption is one of the significant characteristic should be considered within designing of routing protocol. A one of major challenge is to improve the lifespan of network within designing of the homogenous sensor network. This paper deal with a new approach for homogenous sensor network. This method intended to balance the consumption of energy at member node as well as head node of cluster during the data transmission stage. The aim of the intended method to improve energy efficiency and lifespan of network. Keywords – Wireless Sensor network, Energy Consumption, Tree structure, Lifespan of network. I. INTRODUCTION A wireless sensor network (WSNs) are organized by different number of sensor nodes randomly set up in the target area [1]. A wireless sensor network consist of the varied range of application such as military, hospital, planetary investigation and farming [2]. The homogenous sensor networks are the sensor network in which all of sensor node have equal hardware capability and initial battery capacity. A WSN nodes having restricted resources such as limited energy and restricted processing capabilities [3]. The designing of routing protocol involves many challenges. The sensor node contain two constraints that are restricted within supply of energy and bandwidth which are accumulates to decide the sensor network design and management on basis of scalability. In most design of WSN, data transfer is performed from member node of cluster to the head node of respective cluster and then forwards to the base station. The design requirement needs to set up effective data transmission technique order to balance the consumption of energy at member node of the cluster and head node of the cluster [4]. There are many routing protocols have been introduced in WSN through they are lagging within their performance, Hence there is need to work out with new techniques for change in the routing protocol. A tree structure is non-linear structure that can be derived from real tree [5]. A tree structure is usually viewed from the upper side to down with its parent at the top and children on the bottom. A tree is typically accessed from its root as parent node, then through its branches as child nodes. Hence there is need to balance the energy consumption due to the data transmission stage. For this purpose, tree structure is implemented in proposed approach. Each member node of cluster sends their data to the nearest parent node until it reaches to the base station. The next task is to gather the data from member node of cluster at head node in energy efficient way in order to improve the lifespan of the network [7]. Data aggregation is process of merging of data arriving from a verity of sources by removing redundant information and compress them which will helps in balancing the consumption of energy. This paper is intended a routing protocol that is energy balanced clustering approach, which deploying tree structure for homogenous sensor network [8]. This will minimize the amount of energy spend at transmission of data among different sensor node, head node and base station which leads to reduce overhead of re-clustering and helps to improve the data transmission rate and lifespan of network. II. RELATED WORK Research work done by previous author is discussed as below In paper [6], W.R. Heinzelman purpose a clustering based protocol that is Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol. This protocol develops randomized rotation of native cluster’s base stations which simply called as head node of related cluster. The head node consistently issues the energy load between the nodes within the sensor network. In this, head node will be directly send data packets to the base station, which are gathered from member node of cluster. In terms of complex network this approach is not at all times possible. This will cause overhead of re-clustering, more energy consumption at head node. Within [9], the author Kyung Tae Kim chooses a best number of clusters by organizing unique technique which sets up threshold rate that involves the possibility of the best number of head node as well as nodes remaining energy. The proposed scheme introduces the different method to improve the lifetime of the
- 2. www.ijcrset.com 14 | PPage International Journal of Current Research In Science, Engineering and Technology network by the construction of tree in each cluster. In this approach, the base station is placed a distant away from the target area and the cluster head sends data directly to the base station without considering how much distance between base station and head node. In this, there is more energy spent on transmission of data packet from the head node to base station and degrade the lifespan of network. In paper [10], K. T. Kim and H. Y. Youn introduced an enhanced LEACH. For these purpose author use a method in which selects a proxy node which play part of the existing head node of the respective cluster with weak power within one of iteration of communication. For discovery and management of failure of head node, this method depends on healthy nodes. In this scheme, the head node is elected randomly which will make uniform distribution within the network. Therefore, energy spend on head node as well as the base station. In this, there may be possibilities of more energy load on head node will quicklyexpire. According to [11], the author Y. HU, W. Li, Z. Kang discovers the node as head node, which having of more energy with higher possibilities as compared to other nodes. Whenever they interact with the base station then head node are set to make the usage of multi-hop passing as well as routing. In case of this method, if the base station is far away from the head node, then more time and energy spend during transmission of data packet. This leads to degrade the lifespan of network as well as energy efficiency of head node. III. PROPOSED SCHEME 1. Problem Statement Within the homogenous sensor network, nodes which are randomly deployed in target area. However, each of node wish to send their redundant data packet over numerous routes. If these data packets are redundant that will cause an energy deficiency of nodes and the lifespan of the entire network. For this problem, nodes are organized into a set of nodes as cluster. Picking head node of cluster randomly rises the problem in which existing energy of certain head node is lower or having larger distance between the base station and head node will quickly expire. For this purpose, head node is selected based on the highest remaining energy of a node, the number of adjacent nodes of a node, minimum distance between the base station and node [12]. In most of the procedure, all of node sends their data directly to head node without considering how much distance between the member node and head node which may lead to energy deficiency of head node. For this purpose, tree structure is used in which member node sends their data packet to parent neighbor node until it reach to head node and also location based cluster joining in which threshold sensing range is used to reduce the size of cluster. In this way, energy efficiency of head node will improved. Afterwards, head node removes redundant data from accepted data packet, compress them forwards it nearest route node which is also closest to the base station. In order to improve energy efficiency, lifespan of the network, the head node calculates weight rate of its adjacent node. The weight rate is based on the remaining energy of adjacent node, number of adjacent node of the current adjacent node as well as the distance between a node and the base station, the energy consumption of node, count of data packet already received by current node from the adjacent node. In this way, the head node forwards their gathered data to the base station in an energy efficient way within less time. The new energy balanced clustering approach aims at balancing energy consumption of the entire network as well as enhancing life span of network by balancing energy consumption at head node and the member node of the cluster during a data transmission stage. 2. Proposed Approach The intended approach which organize node of the network in the form of tree structure. Within the target zone, this will helps to balance the energy capacity spent on the nodes of the network. The five stages are considered within these proposed approach. 2.2 Deployment of Homogenous Sensor Network In proposed system, sensor nodes are randomly deployed in the network, which holds random topology. All sensor begin with same initial energy. All nodes homogenous and have the identical capabilities. Each node is assigned a unique identifier. The base station is located at the center of the target area. 2.3 Formation of Cluster and Selection of cluster head The homogenous sensor network will be divided into an ideal number of clusters. The design of establishment of cluster is based on tree configuration. Within this stage, some of nodes are randomly selected as originator node. Any of the originator node with greater remaining energy will be playing role of HighEnergy_originator node. Therefore, the originator node will send the association request message to adjacent nodes within two hop distance. Afterwards, all node of homogenous network should accept the association request from originator node in order to form a cluster. The node which belongs to cluster are called as a member node of the cluster.
- 3. www.ijcrset.com 15 | PPage International Journal of Current Research In Science, Engineering and Technology Within next stage, the selection of the head node is implemented. The election of head node lies on greater energy, lowest distance from the base station, threshold rate and remaining energy of node in that cluster. The member node of the cluster becomes an applicant node of head node in election procedure if the remaining energy of node is larger than the threshold rate. The threshold rate can be defined as minimum energy required for accepting data packets from all nodes of the network, gathering them and forwarding that to the adjacent node. Each of applicant node wants to become a head node itself. For these purpose, there is need to calculate these cluster values for each node of the network as C However, C is node m’s cluster value. is node m’s remaining energy. are node m’s neighbor node count. After this calculation, every member node sends its individual cluster value to its neighbor node. Then, each member node of cluster accepts the cluster value of all neighbor nodes in order to distinguish that from its individual cluster value to become a head node of related cluster. The node which contains, the greater cluster value and lowest distance from base station will send an advertising message should send to its neighbor nodes. After receipt of advertising message, its neighbor node decides their neighbor node through considering parameter as received signal strength of different head node. For the purpose of decreasing the cluster size, the threshold sensing range is used. It is expected that the received value of TRs that is based on the sensing area of nodes. Therefore all member nodes of the associated cluster have sent their association request to head node that is closer to them and also send their remaining energy, their locations. After selection of head node, all member nodes are informed about the head node of recent iteration 2.4 Aggregation of data at cluster head Once the formation of the cluster and selection of cluster head is complete then the next stage is aggregation of data at cluster head. The nodes which are present within the sensing range of the particular head node will initiate sensing stage. Therefore, every member node of cluster sends the resultant sensed data to the respective head node of cluster. Therefore, the head node of the particular cluster eliminated the redundant data, gather data and compress the resultant data. 2.5 Data Transmission towards base station The next stage is the data transmission towards base station. Therefore, the gathered data packet from the head node of related cluster, which will be forwarded to the base station in an energy efficient way in less time. For this purpose, the head node will compute a weight rate for all neighbor nodes which have a distance from the base station lesser than itself to send the gathered data [13]. The weight rate of neighbor node m should be computed by node n is specified as However, E is the neighbor node m’s remaining energy. T1 is a constant value. is the number of neighbor nodes of neighbor node n. is the distance between base station and node n. is the node n’s energy spent rate. is node m previously received number of data from node n within the recent iteration. The node will be directly send data packet to the base station if and only if these nodes are nearest neighbors to the base station. For every member node of the cluster transmit their sensed data to its parent node within the assigned period. This process continuous until data transmission of all nods is completed. In proposed system, sensor nodes are randomly deployed in the network, which holds randomtopology. 2.6 Performance Analysis Parameter The performance of the proposed scheme will be evaluated as below
- 4. www.ijcrset.com 16 | PPage International Journal of Current Research In Science, Engineering and Technology 2.6.1 Consumption of energy One of performance measure is the consumption of energy in which energy spend within different procedures such as transferring data packet and electing head node within each iteration. Hence, total consumption of energy can be evaluated as the sum of consumption of energy of the all nodes within the target area. 2.6.2 Packet Delivery Ratio This performance metric is association amongst the total count of data packet accepted by the base station will be distinguished from total count of data packets sent by all nodes of the network. 2.6.3 Throughput Throughput is one of performance metric in which base station accepting a number of data packets. IV. PROPOSED ARCHITECTURE Base Station Cluster Head Cluster Member Routing Route_node Sensing Range Fig 1: Formation of Cluster in Homogenous Sensor Network A group of variety of numbers of sensor nodes which have equal hardware capability and initial energy capacity is termed as homogenous sensor network [1]. These sensor nodes are interacting with each other as well as base station. In order to reach more accuracy over a larger geographical area, an enormous number of nodes permit sensing within the associated target area. Figure 1 shows a representation of diagram of formation of cluster within the homogenous sensor network. In each of the cluster there is one of node elected as a head node using maximum cluster and lowest distance from the base station. The remaining nodes, which are within the sensing area act as member node of that cluster. Afterwards, each member node of cluster sends their sensed data to parent node and so on until that data packet reaches their head node. After receiving sensed data from member node, the head node aggregate data, eliminate the redundant data and compress them [7]. The head node forwards that resultant data to its parent node with highest weight rate and so on until that data packet reaches the base station. In this way, the base station gathers data form all of the head nodes within the target area V. CONCLUSION A Therefore, the selection of head node randomly within a homogenous sensor network causes that the existing energy of a certain head node is lower or their distance from base station are far away. The larger energy load on the head node will quickly expire. For this problem, the new energy balanced clustering method which aims at balancing energy spend on the entire network and prolong lifespan of homogenous network by balancing consumption of energy at head node of respective cluster and member node of cluster during the data transmission stage.
- 5. www.ijcrset.com 17 | PPage International Journal of Current Research In Science, Engineering and Technology REFERENCES [1] Long Zhaohua, Li Jianfeng, Jiang Gujquan, Survey of routing protocols in wireless sensor networks based on application, 2nd International conference on industrial and information system 978-1-4244-8217-71102010 IEEE. [2] Riyazahamed A Jamadari, Mousami S. Vanjale, Threats and attacks analysis of routing protocols for wireless sensor networks, International Journal of Science and Research (IJSR) Vol.3 Issue 12, December. 2014. [3] Shekhar kumar, Pravinash, survey on energy efficient protocol for wireless sensor network, International on conference of computer application (0975-8887), Vol. 120- No.02, June2015. [4] S. Roy, Kumar Das A., energy efficient cluster basd routing protocol (EECBRP) for wireless sensor network, network and soft computing (ICNSC), 2014 first International conference, Page No. 20-29 DOI:1,1109/CNSC.214.690660. [5] M. Lee and V.M.S. Wong, energy aware spanning tree algorithm for data aggregation in wireless sensor networks, IEEE PacRrim 2005. [6] W. R. Heinzelman, A. Chadrakasan and H. Balkrishna, energy-efficient communication protocol for wireless sensor network, In Proc.HICSS, pp.1-10, Jan.2000. [7] Ramesh Rajagpalan and Pramod k. Varshney, data aggregation technique in sensor networks: A Survey, IEEE Communications Surveys & Tutorials 4th Quarter 2006, Vol.8. [8] Mrityunjay, Singh, NiranjanLal, Monika sethi and Saroj Poonia, a tree based routing protocol for mobile sensor networks (MSNs)”, (IJCSE) International Journal on Computer Science and Engineering, Vol.02, No .01S, 2010, 55-60. [9] Kyung Tae Kim, Man Youn Kim, Ji Hyeon Choi, Hee Yong Youn “An Energy Efficient and Optimal Randomized Clustering for Wireless Sensor Networks” IEEE SPND 2015, June1-3-2015. [10] K. T. Kim and H.Y Youn “PEACH: Proxy-Enable Adaptive Clustering Hierarchy for Wireless Sensor Network”, In proc. ICWN, pp. 5257, June 2005. [11] Y. Hu, W. Li, Z. Kang, “Study on Energy Efficient Hierarchical Routing Protocol of Wireless Sensor Network”, In proc. ICIE, pp. 325-328,2009. [12] JaimeLortMauri, Sabu M., Thampi, Danda B. Rawat, Di Jain, “Security in Computing and Communications: Second International Symposium”, Pg.20-29, Sept 2014. [13] Sohoni Roy, Ayankumar Das, “Cluster based Event Driven Routing Protocol Wireless Sensor Network”, International Journal of Computer Application (IJCA), Vol.88, Feb2014.