CiiT International Journal of Networking and Communication Engineering, Vol 8, No 10, November – December 2016 342

An Efficient Greedy Algorithm to Improve the

Lifetime of the Wireless Sensor Network
Dr. J. R. Arunkumar and Dr. M. Sundar Rajan

Abstract---In recent days Wireless Sensor Networks (WSNs) is As shown in the figure 1 the wireless sensor network is a
one of the important area of research to enhance the smart sensors system which is comprised of many sensor nodes that has an
resulting in cost effective and make it compact. In recent trends the ability to interact or sense the surrounding world. It is a process
wireless sensor network plays a major place in transferring the data in of communicating the data with each other and connect the
mobile element. The sensor nodes have platform in the remote
monitoring and control. Several applications were made with
outside world wirelessly. From the architecture it consists of
minimum power nodes and produce the effective communication. This sensing unit, processing unit, transceiver and the power supply
research helps to identify the problem in nodes as well as networks. unit. The data from the sensor is analog so it is necessary to
The proposed design helps to improve the life time of the node. In convert it into digital through the Analog to Digital Converter
mobile network the elements are portable and also battery operated, (ADC) unit. The processing unit is used here to manage the
hence power consumed from the batteries becomes more, in order to communication and data processing functions. Transceiver unit
minimize the power consumption, the proposed work made a new
strategy for each node.
is used to connect the node with each other in a network. For
routing the sensors require location information which is sensed
with high accuracy. The mobilizer needed in some applications,
Keywords---Wireless Sensor Network, Network lifetime, Sensor
Node, Optimization, Greedy Algorithm if the node is required to move to carry out the required tasks.
The battery rating is based on the sensor around 0.5Ah, 1.2v.
I. INTRODUCTION Hence, sensor lifetime is restricted by battery lifetime. The
research is based on the power aware protocols and algorithms
S ENSOR plays a major role in wireless networks, it is used to
monitor and control the physical environment based on the
application. The network consists of large number of sensor
for sensor networks. In traditional approaches the failure of few
nodes can cause relative changes in the topologies and it need
rerouting and reorganization. These failure may occur due to
nodes with close to each other. The working of sensor node is
the draining of power and environmental interfaces. Hence
to measure the parameters which transmit the data collected to
some of the limitation of the low cost sensors are batteries. It is
one or more sinks, using hop-by-hop communication. If the
very difficult to recharge the batteries which is located in the
sinks receive the sensed data then it processes and forwards it
remote environments. To improve the lifetime the waste of
to the users. Sensor is a battery powered, it worked based on the
communication is to be minimized. The WSNs has some
deployment of the nodes and initiation. These sensor nodes are
research challenges due to sensors limited capabilities, some of
usually expected to be functional for several months to years
the challenges addressed below such as self-organizing,
without recharging.
Location discovery, Coverage, Scalability and Proximity. In
wireless sensor network Optimizing energy consumption to
Location Finding Mobilizer increase the network lifetime is important issue. Some of the
System
especial events want to detect and communicates through the
Network lifetime is defined as several types
Sensor / ADC Storage 1. It is the time until the first sensor node in the network
Transceiver runs out of the energy.
2. In another format it is defined as time during which the
Power network is operational.
Generator 3. In some cases lifetime is termed as maximum time
required to operate the node in wireless.
Power Unit
4. Some cases it is termed as maximum duration of time
which deployed the sensors having the ability to
Fig 1: The Components of the Sensor Node. monitor the data.

Manuscript received on October 13, 2016, review completed on October 15, Dr. M. Sundar Rajan is with the Department of Electrical and Computer
2016 and revised on October 25, 2016. Engineering, Arbaminch Institute of Technology, Arbaminch University,
Dr. J. R. Arunkumar is with the Department of Computer Science and Arbaminch, Ethiopia. E-Mail: [email protected]. Mobile: +251-
Information Technology, Arbaminch Institute of Technology, Arbaminch 935360192
University, Arbaminch, Ethiopia. E-Mail: [email protected]. Mobile: Digital Object Identifier: NCE112016003.
+251-932713335

0974-9713/CIIT–IJ-6929/05/$20/$100 © 2016 CiiT Published by the Coimbatore Institute of Information Technology

CiiT International Journal of Networking and Communication Engineering, Vol 8, No 10, November – December 2016 343

The network lifetime is the important metric for battery than that of the processing unit. Hence to improve the lifetime
operated networks. It is also defined by Tang & Xu (2008) [1], some of the routing techniques were concentrated by applying
as amount of time until the sensor runs out of energy. the energy-aware mechanisms to protocols in WSNs. Eghbali
It is given by et al., (2009) [7] mentioned same category of protocols which
is mentioned by Basagni et al., (2009) [8]. The Lifetime
𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝐸𝑛𝑒𝑟𝑔𝑦 𝐸0 improvement mechanisms in protocols were divided into two
𝑇𝑘 = =
𝐸𝑛𝑒𝑟𝑔𝑦 𝑐𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑝𝑒𝑟 𝑢𝑛𝑖𝑡 𝑡𝑖𝑚𝑒 𝐸𝑘 categories such as simultaneous schemes which is mentioned
by the Akkaya & Younis (2005) and cross-layer schemes.
The network lifetime is given by
Gradient-Based Routing Protocol for Load Balancing
𝑇 = 𝑚𝑖𝑛𝑇𝑘
(GLOBAL) is one of the multi-sink protocol that is used to
Several approaches available for maximize the network maximize the network lifetime, it is discussed by Yoo et al.,
lifetime such as Data aggregation, clustering, Topology, (2010) [9] with the help of gradient model. This path selects the
Scheduling, rate allocation, routing and Mobile Relays. The least-loaded path for data forwarding that excludes the
proposed method used to extend the network’s lifetime is by overloaded sensor nodes. By this method the network lifetime
dividing the sensors into a number of sets. Some of the research is not limited by the overloaded nodes. Wang & Wu (2009) [10]
conducted based on the following parameters such as Number presents the Multi-Sink and Load-Balance Routing Algorithm
of sensors in network, Network Radius, Full power capacity of (MSLBR) to increase the network lifetime through distribution
battery, Area to be monitored by wireless sensor network and of loads among deputies. By using this way the individual data
Power consumed by various components of the sensor. packets generated by each source node can select the sink
The remaining part of this work is organized as follows, neighbors randomly to find them. This method is for only load
Section 2 represents the literature survey, next the problem balancing by using round-robin mechanism and according to
identification is stated in section 3, Section 4 explains about the forward factor of neighbors. In order to enhance the network
proposed research methodology with standard technique, lifetime, Luo & Hubaux (2005) [11] proposed a protocol based
Section 5 discuss about the experimental studies and finally, on fixed path/predictable sink mobility and joint routing. They
Section 6 describes about the conclusion of the work. noticed that the maximum lifetime for the sensor network could
only be achieved if the considered sink path covers the
II. LITERATURE SURVEY periphery of the sensor field. The algorithm considered two
patterns for the sink movement while the sensor nodes were
This section describes the traditional methods that describes
distributed in a circular-shape network. In that first path a sink
about the tradeoff between data quality and energy
can move and stop periodically with a discrete mobility
consumption to increase the network lifetime. Tang & Xu
strategy. Another path is based on the help of location
(2008) [1] identified the factors affecting the lifetime of sensor
information as a continuous mobility pattern. To improve the
nodes with respect to the data aggregation such as changing
lifetime Kalpakis et al., (2003) [15] discussed about the
pattern of sensor readings, residual energy of the sensor nodes
adjustable sensing ranges in WSN. In this section several
and communication cost between the sensor nodes and the base
techniques were discussed related to lifetime of the network and
station. Here the results were derived based on the precision
based on that the problem is identified.
location in multi-hop networks. Some of the real data over a
wide range of system configurations is made.
III. PROBLEM STATEMENT
Eslaminejad et al., (2011) [2] listed the categories of routing
protocols for WSNs which is followed by Akkaya & Younis In this section the lifetime of the network is considered as a
(2005) [3]. Several categories made but it is not sufficient to the critical task and analyzed based on the coverage problem of
recent energy-efficient mechanisms such as mobile sink, multi- finding the maximum number of sensors which is covers in
sink, etc. WSNs. Cardei et al., (2005)[12] discussed the Energy-efficient
A taxonomy of several energy-saving techniques applicable target coverage problem to maximize the lifetime of all sensors.
in wireless sensor networks which is developed by Anastasi et For example consider an area of size M×M in which there is a
al., (2009) [4] and Mahfoudh et al., (2008) [5]. Based on these target set T= {t1, t2, t3…, tn} which consists of ‘n’ number of
surveys the energy-aware routing protocols in sensor networks target which are randomly placed in the above specified area.
are separated by considering several factors such as mobility, In order to monitor the targets for maximum possible time, a set
data cycling, topology control and data-driven techniques. of sensor S = {s1… sm} randomly deployed with fix sensing
Since it focused only on the energy steps not on the network range represented as ‘Rs’. Every sensor ‘si’ is assigned an initial
layer hence these researches may not include in multi-sink battery life bi. A target tj, 1 ≤ j ≤ n, is said to be covered if it is
mechanisms for routing protocols. The power failure in sensor falling within the sensing range of at least one sensor si, 1 ≤ i ≤
networks were occur depends on the nodes' characteristics. m. A sensor can be in either active state or sleep state. In the
Raghunathan et al.,(2002) [6] discussed the power properties active state, the sensor can sense information of covered targets
of a Stargate sensor node are different from those called motes and, in the sleep mode it conserves its energy and does not
and also mentioned some of the following components such as monitor the targets. To find out the maximum number of non-
the energy consumption of communication unit is much higher disjoint sets of the sensors called sensor cover, which are of
course upper bounded by Cmax, so that, at a given point of time

0974-9713/CIIT–IJ-6929/05/$20/$100 © 2016 CiiT Published by the Coimbatore Institute of Information Technology

CiiT International Journal of Networking and Communication Engineering, Vol 8, No 10, November – December 2016 344

only one sensor cover is active. Let C be the set of the entire complete problem mentioned in Cardei et al., (2005), but, while
sensor covers, i.e. C = {C1, C2…, Cmax}. Each sensor cover Ck computing the UB on the number of full sensor cover sets has
(Ck ∈ C) is enough to cover all the targets in the network. The to follow the following method:
life time of a sensor cover Ck, X (Ck), cannot exceed the Considering a network of m sensors {s1, s2,…,sm} with their
remaining energy of a sensor in Ck which has the minimum respective energies as {b1, b2,b3,…,bm}and n targets
battery lifetime, represented by X (Ck) = MinsiɛCk(bi). The {t1,t2,t3,…,tn}. The maximum coverage time for target tj denoted
objective of the target coverage problem is to generate by Uj, is given by the following equation.
maximum number of sensor covers to prolong network lifetime.
𝑚

IV. METHODOLOGY 𝑈𝑗 = ∑ 𝑅𝑖𝑗 𝑏𝑖

𝑖=1
1) Energy-Efficient Heuristic Method Where, Rij is the sensor-target coverage relationship matrix.
To maximize the network lifetime, here proposed an energy- Now, the upper bound on the network lifetime, denoted by UB,
efficient heuristic for the given target coverage problem. For is given by the following equation UB = min {U1, U2. . . , Un}.
better understanding of the remaining paper, provided some
required definitions and assumptions. Liang & Liu (2007) [13] Input M, B
Initialize k/Q, max_run, and priority calculated using
mentioned online data gathering for maximizing network
battery power
lifetime in sensor networks. Liang et al.,(2010) [14] discussed
for r = 1 to max_run do
about improving lifetime with mobile sinks.
for iteration = 1 to ∑𝑚𝑖=1 𝑏𝑖 𝑑𝑜
Definition 1: Sensor–target coverage relationship matrix. The if cover possibility exists then
sensor-target coverage relationship matrix Rmxn, is defined as Determine cover based on priority
follows: Optimize cover
1 𝑖𝑓 𝑡ℎ𝑒 𝑠𝑒𝑛𝑠𝑜𝑟 𝑆𝑖 𝑐𝑜𝑣𝑒𝑟𝑠 𝑡𝑎𝑟𝑔𝑒𝑡 𝑡𝑗
𝑅𝑖𝑗 = { Activate optimized cover and reduce battery power
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒 else
Definition 2: Critical target set (T critical) a target tj is critical break
target if sum of the entire sensor’s battery lifetime covering it is end if
minimum among all the targets in T. end for
Calculate network lifetime (nlife)
𝑗 = arg min ∑ 𝑅𝑖𝑗 𝑏𝑖 if nlife < U then
𝑖 Consider weight due to covered targets to compute
priority to
Definition 3: Critical sensor set (Scritical) a sensor si is said to be Check for better lifetime
critical sensor if it covers at least one critical target. else
break
Definition 4: Sensor cover given R, a sensor cover Ck, is a set end if
of rows of R, such that, for every column j, there is a row i in end for
Ck such that Rij = 1. A sensor cover Ck is a minimal cover if for
any cover Ck’, Ck’⊆Ck if and only if Ck’ = Ck.
Algorithm 1 Heuristic for Sensor Node Scheduling Proposed design
Definition 5: Lifetime of a sensor cover The maximum lifetime
based on Greedy algorithm
which can be assigned to a sensor cover Ck, is the smallest
available battery lifetime of sensors which are part of it. Thus
2) Proposed work based on Greedy algorithm
𝑋(𝐶𝑘 ) = min (𝑏𝑖 )
𝑆𝑖 ∈𝐶𝑘
Definition 6: Network lifetime The proposed routing protocol is based on the modification
The total network lifetime (L) can be defined as the sum of all of AOMDV inspired multipath routing protocol with load
the generated sensor covers lifetime balancing and distribution. In this method the paths generated
by disjoint nodes to balance the energy consumption. To avoid
|𝐶|
the long path each packet causing the energy consumption is
𝐿 = ∑ 𝑋(𝐶𝑘 ) avoided. Greedy Algorithm is also used for data assignment.
𝑘=1 The steps which is given below shows the routing protocol for
generating multiple paths.
Definition 7: Upper bound (UB)
The maximum number of non-disjoint sensor covers cannot INPUT: Data for transmission L of a sensor node, whose
exceed the maximum number of full sensor cover sets that routing table has a set of n paths, Pi, to the sink node,
satisfy the coverage constraint. Therefore, the maximum 1 ≦ i ≦ n. Pi has residual energy rei, energy consumption rate
number of full cover sets (Cmax) can be used as the upper limit Ei, bit error rate ßi.
of the number of non-disjoint sensor covers. Finding the OUTPUT: Optimal data chunk size mi for i-th path, 1≦i≦ n.
maximum number of full sensor cover subsets is an NP-

0974-9713/CIIT–IJ-6929/05/$20/$100 © 2016 CiiT Published by the Coimbatore Institute of Information Technology

CiiT International Journal of Networking and Communication Engineering, Vol 8, No 10, November – December 2016 345

Step 1: Sort the paths in a descending order of rei, Where 1 ≤ i ≤ n.

Step 2: Set i=1, where i denotes the number of paths used for data transmission
Step 3: Analyze the value L is equal to 0.
If the answer is positive, terminate.
else go to step 4.
Step 4: Check the value ∑𝑎=𝑛𝑎=𝑡 𝑀𝑡 is less than L, then the algorithm terminates.
else go to step 5.
Step 5: Set mi=Mi, where mi is the maximum allowable number of transmitted bits
If mi is less than or equal to L, then set i=i+1, L=L-mi and repeat STEPs 3 to 5.
else set mi=L, i=i+1,L=L-mi and repeat STEPs 3 to 5.

Algorithm 2 Proposed Greedy Design

The greedy algorithm determines the paths for data chunk

800 752
transmission. Before starting transmission, the Reed-Solomon
code and interleaving technique are applied to the data chunks Energy-Efficient Heuristic Method
700
for communication reliability. The following flowchart 621
Proposed Greedy Algorithm
600

Network Lifetime(ms)
illustrates how Reed-Solomon code and interleaving technique
are used in this paper. By encoding the original data using the
500 453
Reed-Solomon code, all data chunks are then interleaved by
transmitting them over different paths based on the previous 382
400
calculation. After these interleaved data chunks are received by
the sink node, they are merged and decoded their Reed Solomon 300
codes to recover the original data.
200 162
V. EXPERIMENTAL RESULTS 75
100
Network lifetime achieved by the Energy efficient heuristic
and Proposed Greedy algorithm’s Upper Bound on the network 0
for different coverage requirements (K = 1, K = 2 and K = 3) is 1 2 3
shown in figure 2 below to evaluate the performance of the K value
network lifetime with respect to the sensor nodes. It also shows
Fig 2: Comparison of Existing and Proposed Design based on
that when 100 sensor nodes are randomly deployed, k = 3 Lifetime Analysis
condition cannot be satisfied in heuristic. But with the proposed
greedy algorithm, the network can be active for some time. The 16
same is observed for above k = 3. Random deployment yields Energy Efficient heuristic
Network Lifetime Cycle (hrs)

14
zero network lifetime, but the proposed can make the network Proposed Greedy Algorithm
survive for more time. 12
The experimental results shows the relationships between the 10
network lifetime and the initial energy capacity for energy
efficient heuristic and Proposed Greedy algorithm were shown 8
in figure 3. The network lifetime is measured in a cycle that 6
each sensor node transmits 100 bits to the sink node, until the
4
remaining energy of any one sensor node cannot send any
further packet. The results show that the number of network 2
lifetime cycles increases along with the increasing initial energy 0
capacity. 500 100015002000250030003500400045005000
Initialization Energy(Joules)
Fig 3: The Relationships between Numbers of Network Lifetime
Cycle and Energy

0974-9713/CIIT–IJ-6929/05/$20/$100 © 2016 CiiT Published by the Coimbatore Institute of Information Technology

CiiT International Journal of Networking and Communication Engineering, Vol 8, No 10, November – December 2016 346

VI. CONCLUSION
In this research, a novel cross-layer framework is formed
which include a multipath routing protocol and a packet
interleaving technique based on modified heuristic technique.
These were made energy based comparisons on the greedy
algorithm. Here the nodes are distributed to achieve the network
lifetime. From the experimental results it is shown that
proposed techniques follows the traditional method, and shows
the lifetime of network improved at the maximum energy when
compared with the minimum. This method helps to ensure the
communication reliability to avoid packet retransmission. In
future enhancement the time divisional protocols were included
and processed to avoid the retransmission and improve the
efficiency effectively.

REFERENCES
[1] Tang, X., & Xu, J. (2008). Optimizing lifetime for continuous data
aggregation with precision guarantees in wireless sensor networks.
IEEE/ACM Transactions on Networking (TON), 16(4), 904-917.
[2] Eslaminejad, M. R., Gopikrishnan, M., Kayande, D., Kumar, T. S.,
Sookhak, M., Vodithala, S., ... & Tondre, V. S. (2011). A review of
routing mechanisms in wireless sensor networks. International Journal of
Computer Science and Telecommunications, 2(7), 1-9.
[3] Akkaya, K., & Younis, M. (2005). A survey on routing protocols for
wireless sensor networks. Ad hoc networks, 3(3), 325-349.
[4] Anastasi, G., Conti, M., Di Francesco, M., & Passarella, A. (2009).
Energy conservation in wireless sensor networks: A survey. Ad hoc
networks, 7(3), 537-568.
[5] Mahfoudh S., Minet P. Survey of energy efficient strategies in wireless ad
hoc and sensor networks. Proceedings of the Seventh International
Conference on Networking (ICN '08); Cancún, Quintana Roo, Mexico.
13–18 April 2008;
[6] Raghunathan, V., Schurgers, C., Park, S., & Srivastava, M. B. (2002).
Energy-aware wireless microsensor networks. IEEE Signal processing
magazine, 19(2), 40-50.
[7] Eghbali, A. N., Javan, N. T., Dareshoorzadeh, A., & Dehghan, M. (2009,
June). An energy efficient load-balanced multi-sink routing protocol for
wireless sensor networks. In Telecommunications, 2009. ConTEL 2009.
10th International Conference on (pp. 229-234). IEEE.
[8] Basagni, S., Carosi, A., Petrioli, C., & Phillips, C. A. (2009, June).
Heuristics for lifetime maximization in wireless sensor networks with
multiple mobile sinks. In 2009 IEEE International Conference on
Communications (pp. 1-6). IEEE.
[9] Yoo, H., Shim, M., Kim, D., & Kim, K. H. (2010, June). GLOBAL: A
Gradient-based routing protocol for load-balancing in large-scale wireless
sensor networks with multiple sinks. In Computers and Communications
(ISCC), 2010 IEEE Symposium on (pp. 556-562). IEEE.
[10] Wang, C., & Wu, W. (2009, February). A load-balance routing algorithm
for multi-sink wireless sensor networks. In Communication Software and
Networks, 2009. ICCSN'09. International Conference on (pp. 380-384).
IEEE.
[11] Luo, J., & Hubaux, J. P. (2005, March). Joint mobility and routing for
lifetime elongation in wireless sensor networks. In Proceedings IEEE 24th
Annual Joint Conference of the IEEE Computer and Communications
Societies. (Vol. 3, pp. 1735-1746). IEEE.
[12] Cardei, M., Thai, M. T., Li, Y., & Wu, W. (2005, March). Energy-efficient
target coverage in wireless sensor networks. In Proceedings IEEE 24th
Annual Joint Conference of the IEEE Computer and Communications
Societies. (Vol. 3, pp. 1976-1984). IEEE.
[13] Liang, W., & Liu, Y. (2007). Online data gathering for maximizing
network lifetime in sensor networks. IEEE transactions on mobile
computing, 6(1), 2-11.
[14] Liang, W., Luo, J., & Xu, X. (2010, December). Prolonging network
lifetime via a controlled mobile sink in wireless sensor networks. In
Global Telecommunications Conference (GLOBECOM 2010), 2010
IEEE (pp. 1-6). IEEE.
[15] Kalpakis, K., Dasgupta, K., & Namjoshi, P. (2003). Efficient algorithms
for maximum lifetime data gathering and aggregation in wireless sensor
networks. Computer Networks, 42(6), 697-716.

0974-9713/CIIT–IJ-6929/05/$20/$100 © 2016 CiiT Published by the Coimbatore Institute of Information Technology