Energy-Efficient Target Coverage in Wireless Sensor Networks Based on Modified Ant Colony Algorithm

International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC) Vol.1, No.4, December 2010
DOI : 10.5121/ijasuc.2010.1403 29
Energy-Efficient Target Coverage in Wireless
Sensor Networks Based on Modified Ant Colony
Algorithm
1
Salma Begum, 2
Nazma Tara, 3
Sharmin Sultana,
Department of Computer Science & Engineering, University of Dhaka, Dhaka,
Bangladesh
1
jennycsdu@yahoo.com, 2
taranaz2003@yahoo.com, 3
jui_sharmin@yahoo.com,
ABSTRACT
One of the major issues in Target-coverage problem of wireless sensor network is to increase the network
lifetime. This can be solved by selecting minimum working nodes that will cover all the targets. This
paper proposes a completely new method, in which minimum working node is selected by modified Ant
colony Algorithm. Experimental results show that the lever of algorithmic complication is depressed and
the searching time is reduced, and the proposed algorithm outperforms the other algorithm in terms.
KEYWORDS
SCP, ACA, WSN, BS.
1. INTRODUCTION
Wireless sensor networks (WSNs) has become the foundation of a broad range of applications
related to national security, surveillance, military, health care, and environmental monitoring.
Here sensors are deployed to monitor a set of targets. As defined in [1], the goal of target
coverage concept in WSN is to have each location in the physical space of interest within the
sensing range of at least one sensor. To maximize network coverage and to provide a reliable,
energy-efficient monitoring depends on selecting minimum number of sensors in active mode to
cover all the targets. So this power saving technique can be regarded as Set Covering Problem
(SCP). By using modified ant colony algorithm this paper solved the minimum set covering,
designed for a minimum set of nodes where node selection procedure is based on the energy of
each node in a set, provide energy-efficient sensor network. The performance analysis through
simulation results show that the algorithm proposed in this paper selects less working nodes
than the other algorithms and maximize the network lifetime as well.
The rest of the paper is organized as follows. In section 2 we present related works on this topic.
Section 3 defines problem description and section 4 presents the simulation results of the
algorithm and in section 5, we conclude the paper.
2. RELATED WORK
The coverage problems can be classified in the 3 types: (1)area coverage [2, 3, 4, 5], where the
sensors cover an area, (2) point coverage [6, 7, 8],where the sensors cover a set of targets, and
(3) coverage problems that have the objective to determine the maximal support path that
traverses a sensor field [9].An important method for extending the network lifetime for the area
coverage problem is to design a distributed and localized protocol that organizes the sensor

30
nodes in sets. The network activity is organized in rounds, with sensors in the working set
performing the area coverage, while all other sensors are in the dormancy mode. Set formation
is done based on the problem requirements, such as energy-efficiency, area monitoring,
connectivity, etc. Different techniques have been proposed in the literature [2, 3, 4, 5], for
determining the eligibility rule, that is, to select which sensors will be active in the next round.
In [1], to prolong network lifetime, maximum set covering (MSC) was selected. In [10],
selecting the minimum number of nodes in area coverage problem was addressed as set cover
problem (SCP). There are many method of looking for the least set covering. It is the least set
covering problem, which selecting the least number of working node cover the all sampling
points in inspected region. In literature [11], greedy algorithm was used to seek the least set
covering. But, greedy algorithm cannot obtain approximate optimal solution. In [10], area
coverage problem was solved by improved ACA, but that cannot ensure energy-efficiency of
the network. In this paper, we solved energy-efficient target-coverage problem [1] by modified
ACA to select the minimum number of set of nodes to obtain approximately optimal solution.
3. PROBLEM DESCRIPTION
In wireless sensor networks, usually, more sensors are deployed than required to cover a point
or region. One of the major issues in WSN is power scarcity. In order to increase the survival
time of network, optimize control for random deployment of network topology along with
reducing the number of active nodes under the premise of performance in maintaining coverage
is needed. In the target coverage problem, the goal is to maximize the network lifetime of a
power constrained wireless sensor network where sensors are deployed for monitoring a set of
targets with known locations and organized into a number of sets, such that all the targets are
monitored continuously[1]. In this method, the minimum number of sets of sensors is selected
where at least one sensor is BS connected to prolong the network lifetime.
The problem definition is given below:
3.1. Target Connected-Coverage Problem
Let us assume a homogeneous sensor network comprised of N sensors s1, s2,. . . , sN randomly
deployed to cover (monitor) M targets r1, r2,..., rM. Each sensor has an initial energy E where it
consumes some energy per time unit for sensing and for communication purpose. Among the
sensors, some are BS connected, known as reference sensors send the sensed information for
processing. The other ordinary sensors communicate with the BS through the reference sensors.
To prolong the network lifetime, the reference and ordinary sensors are scheduled alternatively
between active and sleep mode such that all the targets are monitored continuously. So, our
target is to select the minimum sets of sensors that would be active for covering all the targets.
3.2. Basic Ant Colony Algorithm for Set Covering
The Set Covering problem is one of the optimization problems. It is usually defined as follows:
Suppose S is a set, S1, S2 ,.., Sm are subsets and coverage of S, namely, ,
solve the minimal covering.
In [12] and others first used ant colony algorithm to solve the SCP. The basic ant colony
algorithm model for set covering is as follows:
At the initial moment, all subsets S1, S2,…, Sm are selected; Ant will be randomly placed on the
m-subsets, assuming that the initial information of each subset is τ j (0) = C (constant). The
probability pij (t) of ant k transfer from the subset i to the subset j is:

31
(1)
Among them, k is the ID( k =1,2,3,..,m )for the ants; t is the iteration number;
representatives the next subset allowed to select for ant k ; said the pheromone strength of
subset j; is the inspired degree of ant k shifted from subset i to the subset j ,this volume is
changeable in the operation system of Ants; These two parameters and , are accumulation
of information and inspired information in the process of ant’s sports, reflects the relative
importance of ants to choice the next subset.
According to equation (1), the inspired degree of ant k shifted from subset i to the subset j
defined as follows:
(2)
Where, M is the number of elements in the S. said the element sets not covered after ant
k select subset i in the cycle t; S j said the number of not covered elements in subset j after ant k
select subset i in the cycle t; After select subset p , Ant will stop when the elements that the
selected subset contained meet , this will mark the end of the cycle. After all the
Ants have completed a cycle, the pheromone of the subsets adjusted according the under
equation:
(3)
(4)
Among them, is the pheromone increment of ants k released in the subset j ; (1- ) is the
attenuation coefficient of the pheromone, usually installed <1 to avoid unlimited accumulation
of informational on subset; J k said the number of subset ant k selected in this cycle; Q is the
pheromone strength, it affected the convergence speed of algorithm in a certain extent.
3.3. Energy-Efficient Target Coverage using ACA
To solve energy-efficient target coverage problem, minimum number of sets of sensors are
required. So target coverage problem is similar to SCP. Since SCP can be solved by ACA,
target coverage problem can be solved by ACA and to make it energy- efficient some
modification on BACA are required. Algorithm steps are as follows:
Step 1 Initially, ants were placed on those subsets where at least one of the sensors are BS
connected. Initialize and .

32
Step 2 Each ant selects the next subset using equation 1. The parameter known as
accumulated information is initialized with default value and the parameter known as inspired
information is defined as follows:
(5)
(6)
Here, c is the number of subset and .E is the energy of each subset i.e. each ant will select the
next subset based on the current energy of each set of sensors.
Step 3 When all the ants will complete one cycle, the pheromone of the subsets adjusted using
the equation 4
Step 4 The iteration will be stopped when the solution set meet the condition of monitoring all
the targets.
4. SIMULATION RESULTS AND ANALYSIS
In this section , the performance of the algorithm are implemented on the PC of memory 1 GB,
Celeron 1.5G by using MATLAB 7.We simulate a stationary network with sensor nodes
deployed in the target region of 100mx100m square region. We also initialize battery energy of
each sensor is 500 mWh and the sensing radius Rs of each node is 20m.
In the simulation, we also consider some changeable parameters such as the number of sensing
node N, number of target nodes M .and the communication radius Rc .We changed the number
of sensing nodes between 20 and 100 and the number of target nodes between 10 and 50. We
also vary the communication range between 80m and 200m.
We take compared test on the basic ant colony algorithm (BACA) and modified ant colony
algorithm (MACA). The default value of ant colony algorithm parameters set to α =.6, ρ=0.9
and Q=30. The main performance metric we pointed out here is the number of working nodes
under different communication range and deployed nodes. We also focused on the number of
coverage ratio determined by different algorithms which indicate the network lifetime of the
WSN.
4.1. Impact on the number of WNs
In the first set of experiments we compare the number of WNs required in WSN under BACA
and MACA. Found through the experiments that approximately 10 to 30 more number of WNs
is required in BACA. Table 1 shows the comparison table and corresponding graph is shown in
Figure 1.
The communication range of sensor nodes impact the number of WNs selected. We compare the
WNs achieved in MACA and BACA in Figure 2.

33
TABLE 1. Comparison table of working nodes under different algorithm
Number of Node WNs in BACA WNs in MACA
20 14 9
30 19 12
40 25 18
50 29 21
60 35 25
70 43 30
80 52 37
90 57 43
100 66 47
Figure 1. Number of WNs selected in BACA and MACA

34
Figure 2. Number of WNs in different communication radius
4.2. Impact on the coverage ratio
The ratio of the target coverage indicates the network lifetime of the WSN. Using the
algorithm in this paper approximately 87% of the coverage can be obtained in the needs
of only 9 WNs among 20 Sensors when targets are fixed as 10 .It is also observable that
sensor with higher density yield more covers. Result is shown in Figure 3.
Figure 3: Coverage ratio in different algorithm

35
5. CONCLUSION
In this paper target connected-coverage problem have been optimized by making it energy-
efficient using modified Ant Colony algorithm. The objective of this paper is to determine the
coverage ratio of all targets which have been significantly improved in comparison with the
Basic Ant colony algorithm. Our future work is to apply the same approach on heterogeneous
wireless sensor network where the energy consumed by each sensor is different.
REFERENCES
[1] Cardei, I, Cardei M. (2008) “Energy-efficient connected-coverage in wireless sensor networks”,
IJSNET, Vol 3. Issue No 3.
[2] Carle, J. and Simplot, D. (2004) “Energy Efficient Area Monitoring by Sensor Networks”, IEEE
Computer, Vol. 37, No. 2, pp. 40-46.
[3] Tian, D. and Georganas, N. D. (2002) “A Coverage Preserving Node Scheduling Scheme for
Large Wireless Sensor Networks”, ACM Workshop on Wireless Sensor Networks and
Applications.
[4] Wang Wang, X., Xing, G., Zhang, Y., Lu, C., Pless, R. and Gill, C. D (2003)”Integrated
Coverage and Connectivity Configuration in Wireless Sensor Networks”, First ACM Conference
on Embedded Networked Sensor Systems.
[5] Zhang, H. and Hou, J. C.(2004) ”Maintaining Sensing Coverage and Connectivity in Large
Sensor Networks”, NSF International Workshop on Theoretical and Algorithmic Aspects of
Sensor, Ad Hoc Wireless and Peer-to-Peer Networks, February.
[6] Cardei, M. and Du, D.-Z. (2005) “Improving Wireless Sensor Network Lifetime through Power
Aware Organization”, ACM Wireless Networks, Vol. 11, No. 3.
[7] Cardei, M., Thai, M., Li, Y. and Wu, W. (2005), “Energy-Efficient Target Coverage in Wireless
Sensor Networks”, IEEE INFOCOM’05.
[8] Cheng, M., Ruan, L. and Wu, W. (2005) ”Achieving Minimum Coverage Breach under
Bandwidth Constraints in Wireless Sensor Networks”, IEEE INFOCOM’05.
[9] Meguerdichian, S., Koushanfar, F., Potkonjak, M. and Srivastava, M. (2001) “Coverage
Problems in Wireless Ad-Hoc Sensor Networks”, IEEE INFOCOM’01, pp. 1380-1387.
[10] Lei Lin, Hou-jun Wang, Zhao Xu. (2008) “Coverage control in wireless sensor network based
on improved ant colony algorithm”. IEEE, Pp 865 - 868.
[11] Chen H, Wu H, Tzeng N. (2004) “Grid-Based approach for working node selection in wireless
sensor networks”. In: Viginier P, ed. Proc. of the Int’l Conf. on Communications. Paris: IEEE
Press.
[12] Ricardo M Silva, A De, Geber L Ramalho. (2001) “Ant system for the set covering problem”.
Systems, Man, and Cybernetics, IEEE International Conference, Vol 5 pp. 3129-3133.

36
Authors
Salma Begum received her M.Sc and B.Sc.
degree in Computer Science and Engineering
from University of Dhaka, Bangladesh, in 2009
and 2007 respectively. Now she is a Lecturer in
Computer Science and Engineering at the
Ahasanullah University of Science and
Technology, Bangladesh. Her research interests
include Ad-hoc Network, Wireless and Sensor
Networking, Digital Image Processing.
Nazma Tara received her M.Sc and B.Sc.
Computer Science and Engineering at The
Peoples’ University of Bangladesh, Bangladesh.
Her research interests include Wireless
Networking, Logic Synthesis and Design,
Reversible Logic, Fuzzy Logic, Multiple-valued
Logic.
Sharmin Sultana received her M.Sc and B.Sc.
Computer Science and Engineering at the
Ahasanullah University of Science and
Technology, Bangladesh. Her research interests
include Networking and Communications,
Mobile Networking, Digital Signal Processing,
Artificial Intelligence and Genetic Algorithm.

Energy-Efficient Target Coverage in Wireless Sensor Networks Based on Modified Ant Colony Algorithm

More Related Content

What's hot (15)

Similar to Energy-Efficient Target Coverage in Wireless Sensor Networks Based on Modified Ant Colony Algorithm (20)

Recently uploaded (9)

Energy-Efficient Target Coverage in Wireless Sensor Networks Based on Modified Ant Colony Algorithm