Energy efficient k target coverage in wireless sensor net-2

International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
254
ENERGY EFFICIENT K-TARGET COVERAGE IN WIRELESS
SENSOR NETWORK
Anurag1
1
(School of Computer Engineering, Kalinga Institute of Industrial Technology/ KIIT
University , KIIT Campus, Patia, Bhubaneshwar, India)
ABSTRACT
Sensing and monitoring a natural phenomenon or target by deploying the sensor
nodes with minimum consumption of energy is a quest for the researchers in designing the
topology of the wireless sensor network. Many proposals have illustrated in the literature,
which are specific in nature and are not optimally feasible, in developing a framework for the
design of appropriate clusters of sensing nodes by considering both their sensing capacity and
energy efficiency. For monitoring any inhospitable environment, sensors are being dropped
randomly from an aircraft and hence they do not fall on precise location. More sensors get
clusters at one location as compare to another. Sometimes the data sensed are being noisy and
the sensors are being vulnerable to failure, hence more number of sensors often required to
cover a particular target. In this paper, we have proposed an energy efficient method to cover
the target with less number of sensor nodes to limit their energy, in which each target has to
be covered by exactly k sensors out of a cluster of n sensor nodes. The collected data passes
to the supervisor node of the respected clusters and supervisor nodes cooperatively send the
data to the sink from there it is available to the user for its further utilization.
Keywords: Dijktra’s algorithm, Energy consumption formula, K-coverage, Relay Node,
Wireless Sensor Network.
1. INTRODUCTION
Wireless Sensor Network is an evolving research field. It has a numerous application
in varieties of fields like civil, military, industrial, agricultural, indoor environmental
monitoring, antique protection [1] and health application[2] to name a few .A WSN usually
consists of large number of sensor nodes for sensing and monitoring the information .This
INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING
& TECHNOLOGY (IJCET)
ISSN 0976 – 6367(Print)
ISSN 0976 – 6375(Online)
Volume 4, Issue 3, May-June (2013), pp. 254-259
© IAEME: www.iaeme.com/ijcet.asp
Journal Impact Factor (2013): 6.1302 (Calculated by GISI)
www.jifactor.com
IJCET
© I A E M E

255
sensor nodes after sensing sends the sensed information to the base station for processing and
from there, it is available to the internet for the end user [3].
One of an important task of the wireless sensor network is the coverage. Coverage deals with
how efficiently a specific region of interest is being covered by the sensors. The goal is to
have the region of interest being monitored by at least one sensor node. The coverage
problem is being broadly classified into the following three types.
1. Target coverage: the goal is to monitor the sets of targets in a region
2. Area coverage: the goal is to monitor the specific area[4].
3. Barrier coverage: the goal is to detect all intrusion in the barrier of the sensor network[5].
To monitor the sets of targets in an area, the sensors are being deployed randomly in
an area. The coverage of a target by a single sensor in not beneficial sometimes because the
sensor are more prone to failure due to harsh environmental condition, energy depletion or
malicious attack[6].Hence it required that a target must be covered by k no. of sensors in
which k≥1. In this paper, we will keep some sensors to sleep mode to save energy to some
extent and hence the target is being covered by exactly k-sensors. We then select a supervisor
node .This supervisor node passes the data to the nearest supervisor node successively, and
finally cooperatively, the data is being transferred to the sink.
Various researches have been done till now related to the target area coverage. In [7],
k-target coverage is being proposed in which a supervisor node is being selected such that it
has the responsibility to monitor the entire k-1 target in an area. This supervisor node
processes the data and sends the result cooperatively to the sink. In [8], the Greedy MSC
Heuristic is being proposed in which the critical target and critical sensor is being selected
which has the responsibility to monitor all the target. This Greedy –MSC proves to be NP-
Complete and has the complexity O(iM2
N), where there are M targets and N sensors. In
[4],an algorithm for the non-disjoint set cover is being proposed in which only those target is
being covered by the sensor which is more closer to the sensor to save the energy to some
considerable extent. In [9],TPISC algorithm is being proposed which works in three phase: In
the first phase, set cover is being constructed which has the responsibility to monitor all the
target. Connectivity to the set –cover is being maintained in the second phase and in the final
stage ,we remove the redundant sensor. In [10],a greedy based algorithm has been proposed
in which a critical target has been selected from the target sets and the unique route from the
sensor set to the sink is being determined by using the Shortest path Tree. This technique
proves to be NP-Complete. In [11], grid based strategies have been proposed in which each
point in a grid must be filled with probability at least T and which the no. of sensors at least
k. In [12],a heuristic algorithm has been proposed which deals with multiple target coverage
problem. There are large number of overlapped target in a region and their corresponding
overlapped sensors. For each overlapped sensor, we have to select the Responsible sensor.
This responsible sensor has the responsibility to transfer the data of overlapped target to the
sink node. RSSA algorithm will then executed in which the number of target observed by
each Overlapped sensor except Responsible sensor has been reduced by one. The complexity
of RSSA algorithm has been given by O(MN) while the complexity of heuristic algorithm has
been given by O(jM2
N) were M and N are targets and sensors respectively.
The rest of the paper is being organized as follows: Section 2 contains our problem
formulation. We describe our proposed work in Section 3 and pseudo code in section 4.
Finally, in section 5, we conclude our paper.

256
2. PROBLEM FORMULATION
We considered the following scenario: large no. of sensors is being deployed in an
area with close proximity of the targets. The sensors are being densely deployed around the
targets. Each sensor has limited battery life and hence is not being rechargeable in most of the
cases. So, energy conservation is being a critical issue in these circumstances. We have the
assumption that the number of sensors deployed in the field is much greater that the optimal
number to perform the sensing tasks. The sensors have uniform sensing range. Each sensor is
aware of the location using localization technique.
A sensor which does the tasks of sensing and monitoring is called source sensor and
sensor which does not perform the sensing tasks and does the tasks of data forwarding are
called relay sensors. The sensor which does sensing and relaying tasks are the active sensors.
Rest of the sensors will be in sleep mode in order to conserve energy.
We assume that all the source sensors have the same data generation rate for a target
i.e. all the source sensors uses the same sampling frequency, quantization, modulation and
coding scheme for each target .Therefore a fixed amount of bit, denoted by β called coverage
rate, is generated by each source sensor for a target in a second.
2.1. Sensing Energy consumption model
The energy consumed in transmitting a bit of data from node i to node j is give by:
Where and b are constants and is the Euclidian distance between i and j.α is the path
loss factor , be the energy consumed in receiving a bit of data and be the energy
consumed in sensing a target for a bit of data. Let be the number of targets a sensor s
can monitors them for t seconds. If Coverage rate is β, the energy consumed in by the source
sensor for t second is the sum of sensing and transmission energy. So, the energy consumed
by the source sensor i which monitors target for t seconds to and transmits the
monitored information to the node j would be:
E(i,t)=β. (t) + i ϵ Ss and i Sr
A relay nodes receives the data from the sensors and transmits them to the other supervisor
nodes .For a given supervising nodes i, let denotes the number of target in which
supervising node i relay the data for seconds. The energy consumed by the sensing nodes is
the sum of sensing energy and transmitting energy. So, the energy consumed in the relay
node i which transfers the traffic to the other relay node j for t seconds would be
E(i,t)= i Ss and i ϵ Sr
A relay node receives data from one relay sensor and transmits them to the other relay sensor.
For a given relay node i, (t) denotes the number of targets which node i relay the data for t
second.
Our proposed model is based upon this method.

257
3. PROPOSED WORK
There is large number of cluster-set for each target, in which each target is being
covered by set of power-constrained sensors. Our aim is to implement an energy–efficient
mechanism for k target coverage in which each targets must be covered by the set of sensors
for longer duration. The network lifetime is being defined as the time in which there exists
any one target which can’t be monitored by any sensor.
Our objective function is to maximize the lifetime of each cluster in the network.
Fig: A k-Target coverage scenario in which k=4
We have use Dijktra’s algorithm to calculate the all pair-shortest path between each
relay node with the complexity of O(|E|).
.It is mainly applicable for the dense network and is
the faster than any other algorithm with the shortest time complexity. The edge between them
represents the energy consumed in transferring the data from one relay node to the other.
A directed graph between the relay sensors pair is being formed in which each edge
represents the energy consumed in transmitting the data.
Our proposed method work as follows: Initially, all the sensors are in sleep mode .We
starts activating all the sensors and the sensors starts covering the target. Large number of
clusters formation takes place for each target. A target is being covered by more than k-
sensors .To save the energy of the sensor; we have to put some of the sensors of lower energy
to the sleep mode in order to conserve energy such that each target is being covered by
exactly k-sensors. The source sensors are already sensing and sending the sensed data by
each source sensor nodes consumes lots of energy, so a sensor called relay sensor is being
selected, this has the responsibility to send the data to the nearest relay sensor of another
cluster. Out of this k -active sensor nodes, we select the sensor with the highest energy as
relay node to save the power to some considerable extent. The Dijktra’s algorithm is being
executed between each relay nodes and the shortest path between each relay nodes gets
obtained. This route has the responsibility to send the data to the sink.

258
When the any of the sensor in the cluster dies, we check the nearest sensor which is in sleep
mode and have higher energy. That sensor will be added to that cluster. Finally, we record the
network-lifetime, target covered and the cluster set C1, C2,, …. , CN.
4. PSEUDO-CODE
1.% Initialization%
2.set l=0;
3.set Ss=S,j=0
4.set l=1;
5.while each target is being covered by at least k-sensor in Ss do
6. % make a cluster set Cj
7.Set j=j+1;Cj=φ and St=T
8. for each tϵSt
9. b(t)≥k
10. for each Si ϵ Cj
11.we keep only the nearest k sensor of higher value of E(i,t) to be in active mode until
b(t)=k;
12.dijktra’s algorithm is executed between each of the k sensors
13.for each cluster Ci, we select the relay node RN which has higher energy
14.compute dijktra’s algorithm between each RN ϵC and find the shortest route between each
relay sensor.
15.for every si ϵCj ,
16. lifetime_si=lifetime_si-w
17. if lifetime_si≤0
18. Cj=Cj-{si}
19. we add the nearest sensor of sleep mode which is of higher value of Energy
20. Cj=Cj U si
21. end if
22 end for
23 end for
24 end for
25 end for
26. end while
27. Return cluster set C1,C2,…,CN and the target covered
Here, b(t) denotes no. of sensors covering a single target . RN is the set of relay
nodes.
5. CONCLUSION
In this paper, we have addressed the k-target coverage problem. It increases the
lifetime of the sensor network to some extent by selecting a supervisor node and transmitting
the sensed data to the nearest supervising node to save energy to considerable extent .We will
implement our proposed work by MATLAB in the next section. We may also apply genetic
algorithm to for its further modification. However, more work has to be done in this field so
that global solution can be achieved.

259
REFERENCES
[1] Yunyue Lin and Qishi Wu,” Approximate Algorithms for Sensor Deployment with k-
coverage in Constrained 3D Space”, in 16th International Conference on Parallel and
Distributed Systems”, 2010
[2] Nahar Sultana, Ki-moon Choi and Eui-nam Huh,” Mobility Support Secure Coverage
Protocol for Monitoring Applications using Wireless Sensor Networks”,in International
Conference on Computational Sciences and Its Applications ICCSA 2008,2008
[3] Purnima Khuntia and Prasant Kumar Pattnaik,” Some Target Coverage Issues of
Wireless Sensor Network”, International Journal of Instrumentation, Control & Automation
(IJICA), Volume 1, Issue 1, 2011
[4] PURNIMA KHUNTIA, PRASANT KUMAR PATTNAIK,” TARGET COVERAGE
MANAGEMENT PROTOCOL FOR WIRELESS SENSOR NETWORK”, Journal of
Theoretical and Applied Information Technology” 15th January 2012. Vol. 35 No.1
[5] Ehsan Saradar Torshizi, Saleh Yousefi and Jamshid Bagherzadeh,” Life Time
Maximization for Connected Target Coverage in Wireless Sensor Networks with Sink
Mobility, in 6'th International Symposium on Telecommunications, 2012
[6] Gao Jun Fan, Feng Liang and ShiYao Jin,” An Efficient Approach for Point Coverage
Problem of Sensor Network”,in International Symposium on Electronic Commerce and
Security, 2008
[7] S.Omid Melli,” K-Target Coverage & Connectivity in Wireless Sensor Network
Considering the angle coverage”,IEEE,2011.
[8] Mihaela Cardei ,My T. Thai ,Yingshu Li and Weili Wu,” Energy-Efficient Target
Coverage in Wireless Sensor Networks”, IEEE INFOCOM 2005
[9]Mohammad ali Jamali, Navid Bakhshivand, Mohammad Easmaeilpour and Davood
Salami,” AN ENERGY –EFFICIENT ALGORITHM FOR CONNECTED TARGET
COVERAGE PROBLEM IN WIRELESS SENSOR NETWORKS”,IEEE,2010.
[10] Yong-hwan Kim,Youn-Hee Han, Chang-min Mun,Chan Yeol Park and Doo-Soon, Park
,”Lifetime Maximization Considering Connectivity and Overlapped Targets in Wireless
Sensor Networks”,IEEE,2010
[11] Wenzheng Zhang and Chuanlin Zhang,” Sensor Placement for Grid Coverage with
Probability Mode” ,IEEE,2010
[12] Sung-Yeop Pyun and Dong-Ho Cho,” Power-Saving Scheduling for Multiple-Target
Coverage in Wireless Sensor Networks”, IEEE COMMUNICATIONS LETTERS, VOL. 13,
NO. 2, FEBRUARY 2009
[13] S.R.Shankar and Dr.G.Kalivarathan, “Feasibility Studies of Wireless Sensor Network
and its Implications”, International Journal of Electrical Engineering & Technology (IJEET),
Volume 4, Issue 2, 2013, pp. 105 - 111, ISSN Print : 0976-6545, ISSN Online: 0976-6553.
[14] Neeraj Tiwari, Rahul Anshumali and Prabal Pratap Singh, “Wireless Sensor Networks:
Limitation, Layerwise Security Threats, Intruder Detection”, International Journal of
Electronics and Communication Engineering & Technology (IJECET), Volume 3, Issue 2,
2012, pp. 22 - 31, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472.
[15] Revathi Venkataraman, K.Sornalakshmi, M.Pushpalatha and T.Rama Rao,
“Implementation of Authentication and Confidentiality in Wireless Sensor Network”,
International Journal of Computer Engineering & Technology (IJCET), Volume 3, Issue 2,
2012, pp. 553 - 560, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

Energy efficient k target coverage in wireless sensor net-2

Related slideshows

More Related Content

Energy efficient k target coverage in wireless sensor net-2