Magnify Network Lifetime in WSN by Reducing Data Aggregation Distance of Weak Nodes

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 6, December 2015
DOI : 10.5121/ijwmn.2015.7602 15
MAGNIFY NETWORK LIFETIME IN WSN BY
REDUCING DATA AGGREGATION DISTANCE OF
WEAK NODES
Prashant Katoch1
and Manik Gupta2
1
Department of Computer Engineering, Chitkara University, Barotiwala
2
District Solan, Himachal Pradesh, India
ABSTRACT
Energy efficient protocols have always played a vital role in conservation of energy in Wireless Sensor
Network. One of the major introduced protocols is LEACH a cluster based protocol. To improve its
performance, an algorithm named as Maximizing the Network Lifetime of Clustered-based WSN Using
Probability of Residual Energy is introduced. This protocol improved the Cluster Head selection process of
LEACH by using the concept of residual energy. In the proposed paper, further improvement is done by
enhancing the data transmission process. This process heightens the lifetime of the nodes having very less
energy left by reducing their data transmission distance. Implemented results in MATLAB shows increase
in the stability and lifetime of the network..
KEYWORDS
WSN, FND, LEACH, modified algorithm, data transmission
1. INTRODUCTION
A collection of large number of wireless sensor nodes having capability of sensing and
transmitting data form a WSN. These nodes having limited capabilities of storage, processing and
power are deployed in a large area at distant places. These nodes collect the data by sensing
process and transmit it further to the base station. Sensor nodes could be of different categories
infrared, radar, seismic and thermal visual [1]. For different environmental and physical
conditions like heat, pressure, temperature surveillance and monitoring of different day to day
activities of our own environment like traffic, home or office monitoring, WSN is used [2] [3].
Nodes have limited power storage and change of batteries is unhandy, so energy and
communication bandwidth are the two major problems that we face [4]. One time transmission of
data by all the nodes to the BS completes one round. With time and rounds energy of the nodes
also starts to decrease. Nodes which are having same energy level in the beginning of the first
round do not have same energy level as the rounds pass, this is because they consume different
amount of energy depending upon the factors like distance and cluster hierarchies [5]. Due to
small battery power and more energy consumption nodes start dying, to avoid it various energy
efficient protocols are utilized. In cluster based approach, some nodes are chosen as CHs and
others work as member nodes [6]. Data collected by member nodes by sensing is trasmitted to
their CHs and CHs will send that data to BS after aggregation [7]. Two types of cluster based

16
routing protocols are used, one is homogenous and other is hetrogeneous. In homogeneous
approach, the sensors nodes are having same capabilities and hardware configurations [8]. In
hetrogeneous approach, sensor nodes are having different capabilities, for example, few nodes are
given high capabilities [8]. In hetrogenity, some nodes have more physical capabilities like higher
bandwidth, can cover longer distance network than others and their batteries are replacable [9].
Our approach in this paper is based on the homogenity. One of the well known approach in
homogenity is Low-Energy Adaptive Clustering Hierarchy(LEACH) [10].. To improve the CH
selection process in this clustering protocol an approach called Maximizing the Network Lifetime
of Cluster-based WSN Using Probability of Residual Energy was introduced [11]. In this paper,
we have introduced an approach to improve the network lifetime and stability of this protocol by
reducing the data transmission distance of the depleting nodes by enhancing the transmission way
of the network.
The rest of the paper is organized as follows. Section 2 describes the related work. Section 3
demonstrates the networking model and working of our protocol. In section 4 the simulation
results are shown and comparison of our protocol with LEACH and MODIFIED ALGORITHM
is performed. In the section 5, paper is concluded.
2. RELATED WORK
2.1. LEACH
Leach is a multiple cluster based protocol in which every cluster has a CH node. All the nodes of
cluster send their sensed data to this node and this node aggregate that data and send to BS hence
reducing the communication overhead of all the nodes.
LEACH is a self-organizing, adaptive clustering protocol that uses randomization to distribute the
energy load evenly among the sensors in the network [10]. With every round CH changes
depending upon the remaining energy because if one node remains the CH then its energy will
refrain quickly. So LEACH is divided into two phases.
One is setup phase, in which CH is selected by the nodes using probability function. Second
phase is steady state phase, in which all the nodes send the data to their respective CH and then
CH send to BS.
Election of the CH is done by selecting a number between 0 and 1 which should be less than the
threshold T(n). Threshold formula T(n) is calculated as [10].
ܶሺ݊ሻ = ൞
‫݌‬
1 − ‫݌‬ሺ‫݀݋݉ ݎ‬
1
‫݌‬
. ݂݅ ݊ ∈ ‫ܩ‬
0 ‫ݐ݋‬ℎ݁‫݁ݏ݅ݓݎ‬
(1)
Where p is the desired number of CHs, r is the current round and G is the set of nodes that has not
been CH in 1/p rounds. Within 1/p rounds, every node will become CH. In the beginning of the
first round, each node has the probability p to become CH and the node which has become CH in
the first round, that node cannot be a CH in next 1/p rounds. Every node’s probability to become
a CH increases. All the nodes would have been CH after 1/p rounds [10].
After the clusters creation and fixation of TDMA schedule the transmission of data packets begin.
Nodes send data to their assigned CHs and at given TDMA slot. Radio of sensor node will only
be on at their respective time slots but receiver of the CH will always remain ON. After the end of
this phase setup phase will begin for the next round.

17
2.2. Modified Algorithm for CH Selection
This algorithm is developed to overcome the limitations of LEACH protocol. In LEACH protocol
CH is selected on the basis of initial energy but after few rounds the energy of the nodes become
uneven then CH selection process is not much of a use. To overcome that disadvantage, this
protocol has introduced the selection of the CH on the basis of node’s residual energy [11]. The
algorithm has been divided in three phases, first is setup phase, second is steady state phase and
third is pre set-up phase.
2.2.1. CH selection
CH selection process involves residual energy of each node. Residual energy is the remaining
energy of a node that can be used for communication in the upcoming rounds. When residual
energy is zero then node is considered as dead. Node’s energy will decrease as the rounds
progress. Residual energy is an efficient parameter to select the CH [12]. Now probability of the
node to become a CH is now depends upon its energy which will help in the better selection of
CH even the uneven residual energies of nodes. So nodes having high residual energy will have
more probability of becoming the CH than the lower residual energy nodes. This helps in
stabilizing the network by ensuring that nodes can die almost at same time. Set the probability of
CH selection as a function of node’s residual energy relative to the aggregate remaining energy
[11]. Threshold energy is calculated as
ܶሺ݊ሻ = ൞
‫݌‬௢௣௧
1 − ‫݌‬௢௣௧ ∗ ሺ‫݀݋݉ ݎ‬
1
‫݌‬௢௣௧
ሻ
.
‫ܧ‬௥௘௦௜ௗ௨௔௟ሺ‫ݐ‬ሻ
‫ܧ‬௧௢௧௔௟ሺ‫ݐ‬ሻ
݂݅ ݊ ∈ ‫ܩ‬
0 ‫ݐ݋‬ℎ݁‫݁ݏ݅ݓݎ‬
(2)
Where popt is calculated as kopt/nr, popt is the CH selection probability, kopt is the optimum number
of clusters in the current round, n is number of nodes and r is number of rounds. G is the set of
nodes which have not been CH last 1/popt rounds, Eresidualis the residual energy of the node and
Etotalis the total energy of the entire network. Etotal(t) is calculated as [11]
‫ܧ‬௧௢௧௔௟ሺ‫ݐ‬ሻ = ෍ ‫ܧ‬௥௘௦௜ௗ௨௔௟
ே
௡ୀଵ
ሺ‫ݐ‬ሻ
(3)
2.2.2. Setup phase
In the beginning of this phase, the probability function pn is randomly calculated and then with its
help threshold value T(n) is calculated. After selection of the CH will make a link with its
associated nodes and TDMA schedule is prepared for all the nodes.
2.2.3. Steady state phase
Its steady state phase is similar to LEACH. In this state the nodes will sense and transmit the data
to its respective CH in its assigned TDMA slot. Every node will send data that it has sensed, its
current energy and its ID. CH then aggregate the all the energy and send to the BS.

18
2.2.4. Next round pre set-up phase
In this phase, the BS receive all the remaining energy level values of all the nodes from their
respective CHs and then calculate the Etotal that is total residual energy of the network and then
send it back to CH. Each node receive this value from their respective CHs and each node save
this value for the calculation of threshold in the next round which completes one round of the
communication [11].
3. PROPOSED PROTOCOL
In this section we will discuss the working of our protocol in WSN. Both LEACH[10] and
modified LEACH[11] algorithms have not provided a way to reduce the energy consumption of
weak or depleting nodes. In this paper, we have discussed a way to reduce the energy
consumption of the weak nodes in the cluster so that they can last for more rounds. It will also
make the network more even and hence the cluster head selection will also be more reasonable. In
the modified algorithm all the nodes in the cluster send their sensed data to their respective CH
and then CH will forward to BS. All the nodes in a cluster excluding CH are treated same despite
of their uneven remaining energies. All the nodes have to transmit the data to CH, so sometimes
nodes are far away from their CHs. This is a problem for the weak nodes as they have to send the
data to a distant place and hence more energy is required for transmission [13]. This leads to the
early death of these nodes. Hence stability of network reduces. Hence to avoid this we have
developed an algorithm to efficiently transmit the data. We introduced a transmission algorithm
in setup phase whereas steady phase and CH selection phase will remain same.
3.1. Setup phase
In the beginning of this round, random probability (pn) by each node is generated and then the
threshold value T(n) is calculated by using node’s residual energy and total residual energy of the
network. Threshold value is calculated as [11]
ܶሺ݊ሻ = ൞
‫݌‬௢௣௧
1 − ‫݌‬௢௣௧ ∗ ሺ‫݀݋݉ ݎ‬
1
‫݌‬௢௣௧
ሻ
.
‫ܧ‬௥௘௦௜ௗ௨௔௟ሺ‫ݐ‬ሻ
‫ܧ‬௧௢௧௔௟ሺ‫ݐ‬ሻ
݂݅ ݊ ∈ ‫ܩ‬
0 ‫ݐ݋‬ℎ݁‫݁ݏ݅ݓݎ‬
(4)
Where Eresidual is the residual energy of the node and Etotal is the aggregation of total energy left in
the network. popt is optimum CH selection probability, kopt is the optimum number of CH and G is
the set of the nodes that never have been cluster CH in last 1/popt rounds. Etotal (t) is calculated
as[11]
‫ܧ‬௧௢௧௔௟ሺ‫ݐ‬ሻ = ෍ ‫ܧ‬௥௘௦௜ௗ௨௔௟
ே
௡ୀଵ
ሺ‫ݐ‬ሻ
(5)
Nodes in the network have same initial energy but after few rounds, energy become uneven and
hence some nodes which are having very less amount of energy left cannot afford a long distant
CH because they have to spend more energy for that. So we have introduced an efficient way for
those nodes to transmit their data. Nodes have been divided into three categories.

19
1) CH node
2) Normal node
3) Weak node
To make sure some nodes in the network do not die early, we have to concentrate on the
diminishing energy of those nodes and have to find a way to stop them dying early as compared
to their companion nodes which will help in making network more stable, cluster making and
easy selection of CHs in network. So, survival of these nodes plays a vital role in the network.
We call these nodes weak nodes. To evaluate which node has to be categorized as weak nodes,
we have built a threshold function.
CH node is selected at the beginning of round using a threshold function. The nodes which are
not CH and have energy more than the weak node threshold function (Tw) can be called as normal
nodes. Weak nodes are those having energy less than the weak node threshold function (Tw) and
hence are near of dying. Weak node threshold function (Tw) can be calculated as
ܶ௪ = ‫ܧ‬௜௡௜௧௜௔௟/10
(6)
Einitial is the initial energy of a node. This equation counts the 10 per cent of the initial energy of
the network. So, nodes having energy less than 10 per cent of the initial node energy are
considered as weak nodes.
3.1.1. Data Transmission Algorithm (CH, DT, BS, WN, NN, N)
1. i=1
2. while I<=N
3. if Ni=CH then
4. Rec(ID, Data)
5. Agg(ID, Data)
6. TransmitToBS(ID, Data)
7. Else if
8. Ni.energy>Einitial/10 then
9. NNi=Ni
10. Rec(ID, Data)
11. TransmitToCH(ID, Data)
12. Else
13. WNi=Ni
14. For (WNi MinDis NNi ║CH)
15. TransmitToNNi ║CH
16. End
Where CH is cluster head, DT is data transmission, WN is weak node, NN is normal node, Einitial
is initial energy of the node, BS is base station, Ni is set of all the nodes.
In this algorithm, first of all we check whether the node is CH or not, if it is, then the data
transmission (DT) will be done to BS. If node is not CH then we will check whether the node has
energy more than 10 per cent of the initial node energy or not. The node which satisfies this
condition will we considered as a normal node (NN) and this node will transmit the data to its
respective cluster head. If node has less than 10 per cent of the initial node energy then the node
is categorized as a weak node and it will transmit the data to the nearest possible node (Ni)
whether it is normal node or CH node.

20
Network stability is very important expect of WSN, nodes of network should not have much
difference in terms of death time. So, this protocol will help the weaker nodes to remain alive for
longer period by reducing their distance of data transmission and hence reducing the amount of
energy utilized. So, all nodes will stay alive for longer period, making the network more efficient
and stable.
A message is then broadcasted to all the nearby nodes and nodes send request message to join the
network and then CH will make a list of its associated nodes. Every node is then given a timeslot
for the transfer of data to CH. That timeslot is called as a TDMA schedule [11] and it will then
broadcasted to all the nodes of the cluster.
3.2. Steady state phase
In this phase, two main tasks are performed. First is sensing and second is transmission of data.
Life of a node also very much depends upon the transmission of data. Energy used in
transmission of data normally depends upon things like distance and packet size[14]. CH transmit
the data in free-space where as a normal node transmit in multipath, so more energy is required if
the data has to be transmitted using free space. We cannot reduce the size of the data but we can
reduce the distance that a node’s data has to cover before reaching its destination. So we have
reduced the distance coverage for the weaker nodes, to increase their lifetime. The weak nodes
will transmit to the nearby node and then that node will transmit data to CH as shown in the setup
phase.
3.3. Next round pre set-up phase
In this round the BS aggregate the residual energy of all the remaining nodes and send back the
value of Etotal to all the CHs[11]. Then CHs will forward it to their respective cluster nodes. This
value will be used in the CH selection process of the next round.
4. SIMULATIONS AND RESULTS OF THE PROTOCOL
4.1. Simulation Parameters
Simulation of this protocol has been done using Matlab simulator. We have obtained the
outcomes of the protocol after certain number of rounds. We have used same parameters on
LEACH, modified algorithm and our protocol to obtain the results. Results have been obtained
the network lifetime using the parameter of First Node Dead (FND). FND defines the number of
rounds after which the nodes start dying. Energy dissipation model which we have used assume
the 50nJ/bit electricity for transmitter and receiver circuitry. To calculate the energy dissipation
radio model has been used[10].
Figure 1. First radio Model[10].

21
4.1.1. First Node Dead
First Node Dead (FND) states that after how many rounds first node in the network is dying. FND
is an efficient way to calculate network lifetime [15]. It shows how stable and efficient a network
is.
Simulation of the protocol in FND shows the better results on both smaller and larger networks in
comparison to both LEACH and modified algorithm. Implementation is done using different
parameters like variation in network size and number of nodes. Simulation parameters are shown
in table 1.
Table 1.Simulation Parameters
Parameters Value
Network Size(M x M)
Base Station
Initial Energy
Packet Size
Nodes
Tx Energy Consumption.
Rx Energy Consumption.
Amplifier energy consumption
(100,100),(500,500)
At the centre of network
.5J/battery/Node
4000 bits
100,75,50
50 nj/bit
50 nj/bit
100 pj/bit/m2

22
4.2. Results and Analysis
Results have been generated and analysed using tables. Flow chart shows the working of
protocol.
Figure.2. Flow chart of the proposed protocol.
Flow chart in Figure 2. shows how the protocol works and how the transmission gets done.

23
4.2.1. Generated Graphs
Graphs have been generated by the implementation of these protocols using different parameters.
Figure.3. FND for nodes=100 and network size 100m x 100m.
Figure.4. FND for nodes=75 and network size 100m x 100m.
Figure.5. FND for nodes=50 and network size 100m x 100m

24
Fig.3, fig.4 and fig.5 represent the enhancement of proposed protocol over LEACH and modified
algorithm in terms of network lifetime of the small scale network (100m x 100m). As the number
of nodes decreases the performance of all the protocols starts enhancing but in all the cases
proposed protocol’s performance over other two protocols is always better.

25
Fig.6, fig.7 and fig.8 shows the improvement of lifetime in proposed protocol over both LEACH
and Modified algorithm for large scale network(500m x 500m) using different number of nodes.
This variation of parameters is applied to show the adaptability of protocol in different
conditions. Improvement in all the three graphs is seen for the proposed protocol. We can see the
exact values of the implementation in the table 2 and 3 given below.
Table 2. Results for 100m x 100m network.
No of nodes 100 75 50
Leach 1309 1398 1382
Modified
Algorithm
1794 1816 1902
Proposed 1919 1998 2187
Table 3. Results for 500m x 500m network.
No of nodes 100 75 50
Leach 29 38 78
Modified
Algorithm
51 67 129
Proposed 79 179 207
We have seen that as the number of nodes increase, the value of FND decreases but still for all
different number of nodes our proposed algorithm shows improvement.
4.2.2. Analysis of results for different parameters
Result shows how parameters different parameters affect differently on the resultant values of the
protocol. As we increase the network size, the network lifetime decreases for all the protocols,
which is because of the fact that nodes have to cover less distance in smaller size networks.
Network lifetime also got affected by the number of nodes. As we decrease the number of nodes,
the network lifetime becomes better. Proposed work shows better results in all the variations of
parameters.
The results show different values for varying parameters, we have seen the results in variation of
two sizes of the network 100m x 100m and 500m x 500m for 100, 75 and 50 nodes. These results
show better percentage of improvement for the larger size network and smaller number of nodes.
But improvement is there in every case. Better results of protocol shows the adaptability of the
protocol in the variation of network size and number of nodes.

26
5. CONCLUSION
Implementation performed on LEACH, modified algorithm and our protocol using a variety of
parameters. We have analysed the results of all of them and have seen that how our protocol
improve the results of network lifetime on the basis of FND over the two protocols. We have
concluded that increasing the lifetime of weak nodes of network can increase the overall network
lifetime of the network. This protocol is much more stable then LEACH and modified algorithm.
Implementation performed in the MATLAB has improved the transmission of data over the
network. This transmission protocol can be applied in many other CH selection algorithms and
help them improving their performance. It has continuously improved the network lifetime for
variety of parameters over modified algorithm and LEACH.
REFERENCES
[1] Sheetal Mehta Parul Bakaraniya, "K-LEACH: An improved LEACH Protocol for," International
Journal of Engineering Trends and Technology, vol. 4, no. 5, may 2013.
[2] A. AbdulMaowjod and H. Mohammed, Q. Ali, ""Simulation and Performance Study of Wireless
Sensor Network (WSN) Using MATLAB," Iraq J. Electrical and Electronic Engineering, vol. vol. 7,
no. no.2, pp. pp. 112 - 119, 2011.
[3] L.F, ""Smart Environments: Tecnologie, Protocols and Application," in Wireless Sensor Network,"
2004.
[4] Anantha P. Chandrakasan,Hari Balakrishnan Wendi B. Heinzelman, "An Application-Specific
Protocol Architecture for Wireless Microsensor Networks," IEEE TRANSACTIONS ON
WIRELESS COMMUNICATIONS, vol. VOL. 1, no. NO. 4, OCTOBER 2002.
[5] Jeremiah D. Deng, Martin K. Purvis Femi A. Aderohunmu, "A Deterministic Energy-efficient
Clustering Protocol," 2011.
[6] Nejah NASRI, Anne WEI, Abdennaceur KACHOURI Salim EL KHEDIRI, "A New Approach for
Clustering in Wireless Sensors Networks," International Workshop on Wireless Networks and Energy
Saving Techniques, pp. 1180-1185, 2014.
[7] Abdalraheem A. Ijjeh Raed M. Bani Hani, "A Survey on LEACH-Based Energy Aware Protocols for
Wireless Sensor Networks," Journal of Communications, vol. Vol. 8, no. No. 3, March 2013.
[8] N. Javaid, A. H. Khan, A. Iqbal, E. Akhtar, M. Ishfaq T. N. Qureshi, "BEENISH: Balanced Energy
Efficient Network Integrated," International Workshop on Body Area Sensor Networks, march 2013.
[9] "HETEROGENEOUS CLUSTER BASED ROUTING PROTOCOLS IN WIRELESS SENSOR
NETWORK- A SURVEY," International Journal of Research in Advent Technology, vol. Volume 1,
no. Issue 4, November 2013.
[10] Anantha Chandrakasan, and Hari Balakrishnan Wendi Rabiner Heinzelman, "Energy-Efficient
Communication Protocol forWireless Microsensor Networks," Proceedings of the 33rd Hawaii
International Conference on System Sciences, 2000.
[11] Saltihie Zeni Jiwa Abdullah, "Maximizing the Network Lifetime of Clustered-based WSN Using
Probability of Residual Energy," IEEE International Conference on Control System,Computing and
Engineering, pp. 28 - 30, November 2014.
[12] D. Q. Lu Cheng and W. Wu, An energy efficient weight-clustering algorithm in wireless sensor
networks.: Proc. Japan-China Joint Workshop on Frontier of Computer Science and Technology,
IEEE Computer Society, 2008.
[13] Kamal Kant, Nitin Gupta Kiran Maraiya, "Wireless Sensor Network: A Review on Data,"
International Journal of Scientific & Engineering Research, vol. 2, no. 4, april 2011.
[14] Doan B.Hoang Najmeh Kamyabpour, "Modeling overall energy consumption in Wireless Sensor
Networks".
[15] Praveen Kaushik, Jyoti Singhai Sourabh Jain, ENERGY EFFICIENT MAXIMUM LIFETIME
ROUTING FOR WIRELESS SENSOR NETWORK. Bhopal, india: International Journal Of
Advanced Smart Sensor Network Systems ( IJASSN ), January 2012, vol. 2.

Magnify Network Lifetime in WSN by Reducing Data Aggregation Distance of Weak Nodes

Recommended

More Related Content

What's hot (15)

Viewers also liked (17)

Similar to Magnify Network Lifetime in WSN by Reducing Data Aggregation Distance of Weak Nodes (20)

Recently uploaded (20)

Magnify Network Lifetime in WSN by Reducing Data Aggregation Distance of Weak Nodes