A new approach for area coverage problem in wireless sensor networks with hybrid particle swarm optimization and differential evolution algorithms

International Journal of Mobile Network Communications & Telematics ( IJMNCT) Vol. 3, No.6, December 2013

A NEW APPROACH FOR AREA COVERAGE
PROBLEM IN WIRELESS SENSOR NETWORKS WITH
HYBRID PARTICLE SWARM OPTIMIZATION AND
DIFFERENTIAL EVOLUTION ALGORITHMS
Isa Maleki1, Seyyed Reza Khaze2, Marjan Mahmoodi Tabrizi3, Ali Bagherinia4
1,2,4

Department of Computer Engineering, Dehdasht Branch, Islamic Azad University,
Dehdasht, Iran
3
Department of Computer Engineering, Science and Research Branch, Islamic Azad
University, West Azerbaijan, Iran

ABSTRACT
One of the most important and basic problems in Wireless Sensor Networks (WSNs) is the coverage
problem. The coverage problem in WSNs causes the security environments is supervised by the existing
sensors in the networks suitably. The importance of coverage in WSNs is so important that is one of the
quality of service parameters. If the sensors do not suitably cover the physical environments they will not
be enough efficient n supervision and controlling. The coverage in WSNs must be in a way that the energy
of the sensors would be the least to increase the lifetime of the network. The other reasons which had
increase the importance of the problem are the topologic changes of the network which are done by the
damage or deletion of some of the sensors and in some cases the network must not lose its coverage. SO, in
this paper we have hybrid the Particle Swarm Optimization (PSO) and Differential Evolution (DE)
algorithms which are the Meta-Heuristic algorithms and have analyzed the area coverage problem in
WSNs. Also a PSO algorithm is implemented to compare the efficiency of the hybrid model in the same
situation. The results of the experiments show that the hybrid algorithm has made more increase in the
lifetime of the network and more optimized use of the energy of the sensors by optimizing the coverage of
the sensors in comparison to PSO.

KEYWORDS
Wireless Sensor Networks, Coverage Problem, Particle Swarm Optimization, Differential Evolution

1. INTRODUCTION
WSNs are used in research, operation and business fields vastly. The WSNs include many sensors
which are applicable in the supervision and security environments [1]. WSNs are able to
supervise the aimed environments and control them and process the gathered information. In
WSNs we must consider the coverage and energy use problems to increase the lifetime of the
network so the data sending and lifetime of the network would not face considerable decrease [2].
The energy use and network coverage are very important factors in designing the WSNs. And
according to the environmental situation of these networks, it is not possible to change the battery
of the thousands of the sensors [3, 4]. So, the coverage problem in WSNs is in direct relation to
the increase of the lifetime of the sensors. The best situation for the WSNs is the time that all
nodes are located in a suitable sensor radius distance. And this means that the network has the
DOI : 10.5121/ijmnct.2013.3606

61


longest lifetime [5]. So, to increase the lifetime of the network, the sensor distribution must be
steady.
The most important factor for developing and scaling the WSNs is to consider the coverage
problem and decreasing the energy use of the sensors [6]. To increase the serious factors in WSNs
it is possible to name these fields [7, 8, and 9]: First the redundant sensors must be deactivated to
save their energy. When the redundant sensors are deactivated, the active sensors gather the
information and send them to the Base Station (BS). It means that if just the active sensors gather
the information, surely the energy use will decrease. So, the coverage must be in a way that all
active sensors be able to overall points and also the minimum number of the sensors will be used
[10]. Second, we must set the coverage area according to the size of the neighbor sensors and then
the distance to the neighbor sensor is needed to sense and transfer the information. So, coverage
problem is the most important and basic case in creation of the WSNs. Different classifications
are identified for coverage in the sensor networks any of which affect the problem from another
point of view.
•

Area Coverage: The most important problem in area coverage is the coverage kind of an
area by the sensors. The main goal of area coverage in WSNs is to cover and supervise en
environment completely [11]. Any point of under coverage environment in area coverage
must be covered at least by one sensor. In [12] the under coverage area is assumed as a
circle and all points under coverage are covered by K sensors. In [13] two K joints and K
coverage states are studied among the sensors and there is no relation between the under
coverage points and the sensors are diffused randomly in the environment. When an area
is covered, any point must be covered by a group of the sensors. So, in this method of
coverage we consider a place coordinates and distribute the sensors randomly or manual
to cover the aimed environment totally. In area coverage the best coverage takes place
when the aimed area is covered by the least number of the sensors completely [14, 15].
Area coverage is often used for the areas in which the probability of events exists in all
coordinates. Also the existence of the redundant sensors in this model cause the multi
coverage which lead to the high density of the network. In Figure (1) the area coverage is
shown.

Figure 1. Area coverage

•

Point Coverage: The goal is to cover a specific point of the environment in this method.
And these points are diffused in the area [16]. It is possible to say that this method is a
sub set of the area coverage method. So, if the total area is not covered by the sensors just
some points are covered and this means the point coverage. In point coverage just the
points which are applicative are covered. So, in point coverage some of the goals are
identified by some specific points which must be controlled are considered. Medium
numbers of the sensors are diffused around the goals and then are activated according to
specific scheduling in relation to specific responsibilities and send the gathered
information to the BS [17]. In Figure (2) the point coverage is shown.
62


Figure 2. Point coverage

•

Barrier Coverage: the total area is not covered by the sensors in barrier coverage. The
barrier coverage is a suitable model for penetration diagnosis applications. In this model
the coverage operations take place in a way that if the penetration takes place from the
width of the under coverage area, the sensors must diagnose them [18]. In barrier
coverage the goals are controlled at least by one sensor and all existing goals in that area
are under control of that area [19, 20]. In Figure (3) the barrier coverage is shown.

Figure 3. Barrier coverage

Now, the population algorithms based on swarm search are used for optimization. An important
class of these algorithms are aspired the natural processes and the behavior of the creatures. These
algorithms are inspired by the swarm behavior of the animals like the insects and are applicative
in optimization problems vastly [21, 22]. In other words the group behavior is used as a powerful
tool for solving the optimization problems. In this paper the problem of area coverage in WSNs is
studied which is one of the most important factors in increasing the lifetime of the network using
the hybrid of PSO and DE algorithms.
This paper is organized as follows: in Section 2, we have introduce the related works; in Section
3, meta-heuristic algorithms are introduced; in Section 4, the proposed model is described; in
Section 5, the evaluation and results of the proposed model are presented and at finally in Section
6, we have presented the conclusion and future works to be done.

2. RELATED WORKS
One of the most important research fields in wireless communications, is WSNs. WSNs includes
a set of the sensors which are diffused in supervising environments and process the sensed data
and finally send the favorite information to the BS. Many different methods are presented for
improving the coverage problem till now any of which has caused many advances in the coverage
and its quality in the WSNs.
Using the Multi-Objective Particle Swarm Optimization (MPSO) in coverage problem for energy
efficiency and the increase of the lifetime of the WSNs is studied in [23]. The main goal of
MPSO is to find the best position for the sensors for better coverage. According to the results of
63


the experiments it is possible to say that MPSO is efficient in coverage and increasing the lifetime
of the network. In [24] studied the K-coverage problem in WSNs using the Harmony Search
(HS). In this reference it is said that the two coverage and energy efficiency are very important
parameters in wireless sensor nodes distribution. In this reference the HS algorithm is used for
better joint of the sensors, K-coverage and minimizing the energy use. In K-coverage the area is
covered by k sensors. The results of the experiments in this reference show that the HS algorithm
is very efficient in coverage and minimizing the energy use of the sensors. In [25] a distributed
algorithm with efficient energy use and optimized coverage goal is proposed. In proposed
algorithm the sensors are divided into active and inactive subsets and create a graph model which
leads to the balance energy use of the sensors. In this reference, only the active sensors are used
for coverage and the rest of the sensors go deactivated. Turning off the unnecessary and
redundant sensors, the other sensors of the network use less energy and as a result the lifetime of
the network increases. The results of the simulations show that the efficiency of the suggested
algorithm is more than the under comparison algorithms and it is able to increase the lifetime of a
network up to multi times more.
In [26] has studied the positioning method of the sensors in WSNs according to the point
coverage. In point coverage which is shaped by some goal points, the establishment of the sensors
is very affective on the number of the needed sensors for coverage of the points. In this reference,
the point coverage takes place according to the position of the goal and angle of the goal points.
Z. Bin et al [27] use the Fish Swarm (FS) and PSO algorithms hybrid to study the WSNs
coverage problem. PSO algorithm is used in hybrid algorithm for more efficiency and FS for
covering the sensors. The results show that the hybrid algorithm is efficient enough in
deployment of the sensors of network and has improved the coverage problem. In some
researches [28], the sensing coverage of any sensor is set in a specified space for increasing the
efficiency of the energy of the sensors. In this reference, two heuristic algorithms are used for the
results of the experiences. The results of the experiments show that Greedy is more efficient in
energy consumption and network coverage in comparison to Linear Programming. The researches
of [29] proposed a heuristic algorithm to fully cover a region having arbitrary (opaque) obstacles,
which allow neither the sensor to be placed inside nor the signals to pass through. They first
deploy an optimal pattern for covering a plane over the region, and then locate and efficiently
cover the uncovered holes formed by the obstacles.
Reference [30] has presented a dynamic algorithm for area coverage problem in WSNs. In
coverage, the most important challenge is the distribution of the sensors and the lifetime of the
network. In this reference, the sensors are distributed according to the supervision area and the
shape the most important establishment. Also the sensors can dynamically have topological
changes in the environment. The most important specifications considered in this reference is the
energy used by the sensors. The results of the experiments show that the proposed algorithm is
more efficient than the other algorithms from establishment of the sensors point of view.
Researchers [31], use the Ant Colony Optimization (ACO) Algorithm in WSNs coverage
problem. The best radius for the sensors is identified using the ACO algorithm. When it is needed
to have an active network for long time and the sensing limits of the any sensor is identified or
when the quality and stability of supervision are the most important factors, the use of ACO
algorithm is suitable. They have evaluated the improvement of the energy use problem in sensor
networks using the area coverage and have showed that ACO is efficient in lifetime of the
network. According to the results of their simulation, ACO algorithm is efficient in lifetime of the
network.
Researchers [32] have used the heuristic algorithms for the coverage problem in large scales. For
optimization of coverage problem in large scales, the Greedy algorithm is utilized. The goal of
them is to use the heuristic algorithms to establish the sensors in the points of the spaces which
cover the network in the best manner and have the best lifetime. Any point could have many
redundant sensors and if they are active, more energy is used. So, the sensors which are not used
64


are deactivated by the Greedy algorithm a just the sensors which are in relation to the base station
are activated. The results of the experiments show that the Greedy algorithm is the best solution
for coverage problem of the sensors in large scale. X. Wang et al [33] have proposed a new
protocol based on the dynamic structure of the sensors for coverage problem. In the proposed
protocol, the sensors are activated or deactivated for energy saving. Also, the position of the
sensors is dynamically identified and any sensor can cover its radius well. They have proved that
the propose algorithm is more efficient in coverage. In [34] first studied the barrier coverage of
two-dimensional plane and two-dimensional strip sensor networks using percolation theory
results. The barrier coverage of a two-dimensional plane network is related to the existence of a
giant sensor cluster that percolates the network. However, the strength of the barrier coverage,
i.e., the number of disjoint barriers, was not obtained. For a two-dimensional strip network of
finite width, it is proved that there always exist crossing paths along which an intruder can cross
the strip undetected. Furthermore, the probability that an intruder is detected when crossing a strip
is characterized.

3. META-HEURISTIC ALGORITHMS
Meta-Heuristic algorithms are based on population to the optimization problems and are tools for
finding the near to optimized solutions. These algorithms utilize the diversity and cooperation
concepts and make the optimization space better in achieving the most optimized status. So, the
more the power of an algorithm in controlling the two parameters, the more algorithms is capable
in finding the most optimized stage. In this section we talk about PSO and DE algorithms which
are the most important population algorithms.

3.1. Particle Swarm Optimization
PSO algorithm was first introduced Kennedy and Eberhart in 1995 inspiring the social behavior
of the birds living in large and small groups [35]. PSO is a simulation of the social behavior of the
birds which search for food in and environment. None of the birds have information about the
place of the food but they know in each stage how far they are from the food. On this basis, the
best procedure to find food is to follow the nearest bird to the food.
PSO algorithm is a population algorithm in which a number of the particles which are the
solutions for a function or problem shape a population. A population of the particles moves in the
problem space and tries to find the most optimized answer in the searching space according to
their own experience and also the population’s. PSO algorithm is an optimization algorithm
which provides a search based on the population in which any particle changes its position by the
time. In PSO algorithm the particles move in a searching space of multi dimension including the
possible solutions. In this space an evaluation factor is defined and the quality evaluation of the
solutions of the problem takes place by it. Any change of a particle in a group is affected by the
self or other’s experience and the searching behavior of a particle is affected by the other
particles. This simple behavior causes finding optimized areas of the searching space. So, in PSO
algorithm, any particle which finds the optimized situation, informs the other particles in a
suitable manner and any particle decides for the cost function according to the achieved values
and searching takes place using the ex-knowledge of the particles. This cause the particles do not
get near each other more than the normal and solve the optimization problem effectively.
In PSO algorithm, first the group members are created randomly in problem space and the
searching process for the optimized answer starts. In the total structure, the search of any member
follows the other which is the most optimized suitable value of the function and it does not forget
its experience and follows the state in which he suitability function value was the most for itself.
So, in each repetition, any member changes its situation according to the two values, one of them
is the best situation of the member till then (pbest) and the other is the best situation the total
65


population has had till then. In fact it is the pbest in the total population (gbest). In concept, pbest
for any member is in fact the biologic memory of the member. gbest is the general knowledge of
the population and when the members change their situation according to gbest, in fact they try to
heighten the knowledge level up to the general knowledge of the population. From the concept
view, the best particle of the group relates all other particles to each other. The identification of
the next situation of any particle is done by the equations (1) and (2).

vi +1 = w.vi + c1 .r1 .( Pbesti − xi ) + c2 .r2 .( g besti − xi ),

( 1)

xi +1 = xi + vi +1
( 2)
In equation (1), c1 and c2 are the learning parameters. rand() is a function for producing the
random numbers in [0, 1]. xi is the present situation and vi is the moving speed of the members. W
is a control parameter which controls the present speed (vi) with the next one and creates a
balanced state between the ability of the algorithm in local and global searching and then reaches
the answer in a shorter time. So, for optimized operation of the algorithm in the searching space,
parameter w is introduced as follows [36, 37]:

w = wMax −

((wMax − wMin ) × i)
iMax
( 3)

In equation (3), imax shows the maximum number of the repetitions of the algorithm and parameter
i is the counter of the repetitions of finding optimized answer. In equation (3), the wmax and wmin
are the primary value and the final value of inertia weight in algorithm execution time,
respectively. Inertia weight value changes from 0.9 to 0.4 linearly in execution time of the
program. Large values of w lead to the global search and the small values lead to local search. To
balance the local and global search it is necessary to reduce the inertia weight in algorithm
execution time. So, by reducing the w, the search takes place locally and around the optimized
answer.

3.2. Differential Evolution
DE algorithm was innovated in 1995 by Storn and Price [38]. DE algorithm is a probable
searching which is based on population. This algorithm uses the distance and direction from the
existing population information to continue searching. He advantages of this algorithm are speed,
parameter setting, efficiency in finding the optimized solution, parallelization, high care and no
need for ordering and matrix coefficient. DE algorithm is able to search in direction of the
coordinates of the optimization variables and changing the coordinates for finding the optimized
solution. DE algorithm starts the evolution of the searching process from a random initial
population. Three mutation, crossover and selection controlling parameters include the
population, index coefficient and the crossover probability which are important in DE Algorithm.
DE algorithm stages include the followings:
•

Creating initial population: The initial population or the solution vectors are selected
randomly from the problem domain in DE algorithm. The solutions position vector is
introduced by equation (4).

X i = ( xi ,1 , xi , 2 ,..., xi , D )
( 4)
66

min
max
min
xik = xk + rand (0,1).( xk − xk )

with

i ∈ [1, Np], k ∈ [1, D]

( 5)
The selection of the random numbers xik from the problem domain takes place by equation (5). In
equation (5), D means the dimensions of the solutions. Np is the number of the initial population.
Rand (0, 1) function produces the random numbers (uniform distribution) in (0, 1). It is clear that
if the equation (5) is used, the values for xik is in [ ximax , ximin ] and the position vector of the
solutions would be one of the potential answers of the optimization problem.

•

Mutation: In mutation stage three vectors are selected randomly which differ each other.
For any x vector in population a new answer in each repetition is created according to
equation (6).

vi ,G+1 = xr1 ,G + F .( xr2 ,G − xr3 ,G )
( 6)
In equation (6), r1, r2 and r3 which are three non-equal random numbers are located in [1, Np]. G is
the number of the produced generations and F is a constant and real number which is often
considered 0.5.
•

Crossover: Crossover operator causes increase in diversity of the population. This
operator is similar to the crossover operator in genetic algorithm [39]. In this operator the
new vectors are created by hybrid of the x and v vectors as equation (7).

v ji , G +1 if (rj ≤ CR) or j = jrand
u ji ,G +1 = 
 x ji, G +1 otherwise
( 7)
In equation (7), parameter CR is located in [0, 1). Parameter rj is randomly created in [0, 1]. Also
the value is j=1, 2 … D.
•

Selection: To select the vectors of the highest propriety, the vectors created by the
mutation and crossover operators are compared to each other and any of them holding
more suitability is transferred to the next generation. The selection operator takes place
by equation (8).

xi ,G +1 = Fitness Value(ui ,G +1 , xi ,G )
( 8)
•

Stop: The searching process continues till the stopping factor of the algorithm is met.
Usually the stopping facto of the algorithm could be based on the non-changing propriety
of the best answer or algorithm repetition.

4. PROPOSED MODEL
The most important factor in WSNs is to minimize the energy use of the sensors and the coverage
problem. Sensors usually use a battery for providing the energy which is not chargeable or
changeable in many cases. So, the reduction of energy use in increase of the lifetime of the
sensors is very important. In coverage problem, the sensing radius of the sensors which shows the
capability of diagnosing the phenomena or physical signals by the sensors is very important.
67


Setting the parameter of sensing radius of the sensors is very important but the minimization of
the energy used by the nodes is also very important. Of the problems to be cited in WSN is the
distribution type of the sensors in an optimized coverage. In this paper a new hybrid model using
the PSO and DE algorithms for distribution of the sensors for area coverage is proposed. In
Figure (4) the hybrid algorithm flowchart is showed.

Figure 4. The hybrid algorithm flowchart

68


In Figure (5), the quasi code of the hybrid algorithm is shown.
1. Initialization Phase
1.1. Start
1.2. Initialize PSO & DE parameters
2. Coverage Phase
2.1. Loop
2.2. Run program for coverage
2.3. Obtains best coverage
●Randomly initialize all particle velocities
●for all particles
●Update V
●Update X
●Compute V and X of each particle
●Update gbest
●Update pbest
●Modification of each search particles
●Mutation vector particles
●Crossover vector particles
●Selection vector particles
2.4. Assign X & Y Coordinate to each sensor
2.5. Sensor (i) = Optimum Location (Xi, Yi)
2.6. Computes the Sensor Position
2.7. Evaluate Object Function for each sensor
2.8. While (Max Iteration)

Figure 5. The quasi code of the hybrid algorithm

In hybrid algorithm the most optimized points for n sensors are searched according to PSO
algorithm for covering p aimed points. Searching takes place based on the particles which are
near the goal and are able to cover the around suitably and then the mutation and crossover
operations on n sensor take place using DE algorithm. One of the most important goals of
mutation operation and crossover is finding more optimized points for the sensors. The hybrid
algorithm acts for increasing the lifetime of the network and reducing the energy use and covering
all points in an optimized manner. The particles in PSO algorithm is affected by the situation in
the neighbor. In hybrid model any particle has a two dimensional vector of x and y coordinates
which is in relation to the neighbor particles. In hybrid model any particle is longing for following
the best point in its neighborhood. In Figure (6) the position vector of the sensors is shown using
the PSO algorithm.

Figure 6. The position vector of the sensors

One of the most important parts in the hybrid algorithm is the evaluation of goal function. In goal
function evaluation the most important parameter is the overlap rate of the sensing radius of the
sensors. The sensing radius of the sensors must be set in a manner to create the best coverage. For
sensing radius coverage of two sensors in p1 and p2 the equation (9) is used. Equation (9) is
updated in each repetition using PSO algorithm and the most optimized points are found
according to the global knowledge.

d ( s1 , s2 ) = ( xp1 − xp2 ) 2 + ( yp1 − yp2 ) 2
(9)
In equation (9), if the sensors are located in non-optimized points, the position of the particles is
updated in hybrid algorithm and the most optimized points are found for the sensors. And the
69


sensors coverage is repeated till the optimized suitability is achieved. To calculate the energy use
of the sensors in this paper the [40] model is used. In this model the value of used energy for
receiving k bits of information is introduced by equation (10).

Rx = ε elec × k
(10)
The energy used for sending one message from a sensor to another sensor located in distance d, is
introduced by equation (11).

Tx = ε elec × k + ε amp × d 2 × k
In equations (10) and (11) the value of ε elec and ε amp are 50nJ

and 100 pJ

(11)
respectively.

bit
bit × m2
In Figure (7), simulation of area coverage problem in WSNs using the hybrid of the PSO and DE
algorithms in C#.NET 2008 programming environment is shown.

Figure 7. Simulation of area coverage problem in C#.NET 2008

5. EVALUATION AND RESULTS
In this section the coverage problem in WSNs which is introduced as one of the quality factors is
studied using the hybrid algorithm of PSO and DE. And also the sensing radius problem in
sensing coverage of the sensors and the lifetime of the sensors are studied. Simulation has taken
place in a 450*450 meter place. In hybrid algorithm there are many parameters which affect the
operation of the algorithm. In Table (1), P parameter is the number of the population (particles).
Parameter W is the inertia weight to balance the speed of the particles and the value of the C1 and
C2 parameters contributes the particles’ learning in finding the optimized points. Parameter F is a
constant positive and real number which is used for convergence rate of mutation. Pm parameter is
the rate of mutation, and parameter Pc is the crossover rate. Also the initial population in PSO and
hybrid algorithm is considered 1000.
70


Table 1. The value of the parameters

Parameters
P
C1
C2
W
F
Pm
Pc

Value
100
1.5
1.5
0.4
0.5
10
0.3

5.1. The Effect of the Number of the Sensors on the Lifetime of the Network
In this section the effect of the number of the sensors on the lifetime of the network is studied.
The radius area of any sensor is considered as 40. Figure (8) shows the effect of the number of the
sensors on the lifetime of the network is studied. As it is clear from Figure (8), the lifetime of the
network in hybrid algorithm linearly increases by the size of the networks.

Figure 8. Evaluation of lifetime of the network

5.2. The Effect the Sensing Radius of the Sensors on the Lifetime of the Network
In this section the effects of the sensing radius area of the sensors on the lifetime of the network
are evaluated. 50 sensors are considered for the results of the simulation. As it could be seen in
Figure (9), the hybrid algorithm is more efficient on the lifetime of the network. Because the
hybrid algorithm sets the radius of any sensor better and causes less energy waste in the coverage
area.

71


Figure 9. The effects of the sensing radius area of the sensors

If the distance of two sensors is high, more energy is used for the information transfer between
them. So, as it can be seen, the more sensors cover each other, the less the distance between the
two sensors and finally the less the used energy. The results of the simulations show that the
hybrid algorithm acts better than the PSO algorithm from convergence speed and reaching the
optimized answer points of view. The hybrid algorithm uses the mutation and crossover
parameters for better coverage for the sensors.

5.3. The Comparison of the Sensor Coverage Percent
Figure (10) shows the coverage percent comparison of the hybrid and PSO algorithms. As can be
seen in Figure (10), as the hybrid algorithm is more able to set the sensing area of the sensors, it
covers more efficient.

Figure 10. The comparison of the efficiency of the coverage in hybrid algorithm and PSO

6. CONCLUSION AND FUTURE WORKS
Coverage problem is one of the most important research fields of WSNs. Creating the optimized
coverage in WSNs it is possible to increase the lifetime of the network. In this paper we have
used a hybrid of the PSO and DE algorithms for area coverage in WSNs and to increase the
lifetime of the network. For this reason in hybrid algorithm the two factors of suitable distribution
of the sensors and the energy decrease which lead to increasing the lifetime of the network use are
considered. And to show the efficiency of the hybrid algorithm better, it is compared to PSO
72


algorithm and according to the results of the simulations, it can be said that the hybrid algorithm
is better. By this paper we hope that we will be able to find better solutions and more optimized
answers using other meta-heuristic algorithms for WSNs coverage problem.

REFERENCES
[1] I.F. Akyildiz, W. Su, Y. Sankarasubrmaniam, E. Cayirci, “A Survey on Sensor Networks”, IEEE
Communication Magazine, Vol. 40, Issue 8, pp. 102-114, IEEE, Aug 2002.
[2] M. Cardei, D. MacCallum, X. Cheng, M. Min, X. Jia, D. Li, D.Z. Du, “Wireless Sensor Networks with
Energy Efficient Organization,” Journal of Interconnection Networks, Vol. 3, pp. 213-229, March
2000.
[3] J. Yick, B. Mukherjee, D. Ghosal, “Wireless Sensor Network Survey,” Computer Networks, Vol. 52,
No. 12, pp. 2292-2330, 2008.
[4] C. Zhu, C. Zheng, L. Shu, G. Han, “A Survey on Coverage and Connectivity Issues in Wireless Sensor
Networks”, Journal of Network and Computer Applications, Vol. 35, Issue 2, pp. 619-632, Elsevier,
2012.
[5] E. Jain, Q. Liang, “Sensor Placement and Lifetime of Wireless Sensor Networks: Theory and
Performance Analysis”, In Proceedings of the IEEE Global Telecommunications Conference
(GLOBECOM ’05), USA, Vol. 1, pp. 173-177, IEEE, December 2005.
[6] Z. Liu, “Maximizing Network Lifetime for Target Coverage Problem in Heterogeneous Wireless
Sensor Networks”, In Proceedings of the 3rd International Conference on Mobile Ad-Hoc and Sensor
Networks (MSN ’07), pp. 457-468, Beijing, China, December 2007.
[7] X.Q. Wang, S.Q. Zhang, “Research on Efficient Coverage Problem of Node in Wireless Sensor
Networks”, International Conference on Mechatronics and Automation (ICIMA), Chengdu, pp. 9-13,
IEEE, 2009.
[8] X. Wang, F. Sun, X. Kong, “Research on Optimal Coverage Problem of Wireless Sensor Networks”,
International Conference on Communications and Mobile Computing, Yunnan, Vol. 1, pp. 548- 551,
IEEE, 2009.
[9] Z. Zhou, Z. Huang, Q. Liu, Y. Zhou, “A Study on the Coverage Problem in Wireless Sensor Network”,
International Conference on Wireless, Mobile and Multimedia Networks, Hangzhou, China, pp. 1-3,
IEEE, 2006.
[10] S. Slijepcevic, M. Potkonjak, “Power Efficient Organization of Wireless Sensor Networks”,
International Conference on Communications, Helsinki, Vol. 2, pp. 472-476, IEEE, 2001.
[11] H. Tan, X. Hao, Y. Wang, F. C.M. Lau, Y. Lv, “An Approximate Approach for Area Coverage in
Wireless Sensor Networks”, Procedia Computer Science, Vol. 19, pp. 240-247, Elsevier, 2013.
[12] C.F. Huang, Y.C. Tseng, “The Coverage Problem in Wireless Sensor Networks”, In International
Workshop on Wireless Sensor Networks and Applications, San Diego, California, USA, pp. 115-121,
2003.
[13] C.F. Huang, Y.C. Tseng, H.L. Wu, “Distributed Protocols for Ensuring both Coverage and
Connectivity of a Wireless Sensor Network”, In ACM Transaction on Sensor Networks, Vol. 3, No. 5,
March 2007.
[14] Kh.M. Alam, J. Kamruzzaman, G. Karmakar, M. Murshed, A.K.M. Azad, “QoS Support in Event
Detection in WSN through Optimal k-Coverage”, Procedia Computer Science, Vol. 4, pp. 499-507,
Elsevier, 2011.
[15] J. Yu, X. Deng, D. Yu, G. Wang, X. Gu, “CWSC: Connected K-coverage Working Sets Construction
Algorithm in Wireless Sensor Networks”, International Journal of Electronics and Communications,
Vol. 67, Issue 11, pp. 937-946, Elsevier, 2013.
[16] M. Cardei, M.T. Thai, Y. Li, W. Wu, “Energy-Efficient Target Coverage in Wireless Sensor
Networks”, 24th Annual Joint Conference of the IEEE Computer and Communications Societies,
IEEE, Vol. 3, pp. 1976-1984, 2005.
[17] S. Yang, F. Dai, M. Cardei, J. Wu, F. Patterson, “On Connected Multiple Point Coverage in Wireless
Sensor Networks”, International Journal of Wireless Information Networks, Vol. 13, No. 4, pp. 289301, October 2006.
[18] A. Chen, S. Kumar, T.H. Lai, “Local Barrier Coverage in Wireless Sensor Networks”, IEEE
Transactions on Mobile Computing, Vol. 9, Issue 4, pp. 491-504, 2010.
[19] S. Kumar, T.H. Lai, A. Arora, “Barrier Coverage with Wireless Sensors”, Proceedings ACM
MobiCom, Cologne, Germany, 2005.
73


[20] B. Liu, O. Dousse, J. Wang, A. Saipulla, “Strong Barrier Coverage of Wireless Sensor Networks”,
Proceedings ACM MobiHoc, Hong Kong SAR, China, 2008.
[21] F.S. Gharehchopogh, I. Maleki, S.R. Khaze, “A New Optimization Method for Dynamic Travelling
Salesman Problem with Hybrid Ant Colony Optimization Algorithm and Particle Swarm
Optimization”, International Journal of Advanced Research in Computer Engineering & Technology
(IJARCET), Vol. 2, Issue 2, pp. 352-358, February 2013.
[22] F.S. Gharehchopogh, I. Maleki, B. Zebardast, “A New Solutions for Continuous Optimization
Functions by using Bacterial Foraging Optimization and Particle Swarm Optimization Algorithms”,
Elixir International Journal Computer Science and Engineering (Elixir Comp. Sci. & Engg.), Vol. 61,
pp. 16655-16661, July 2013.
[23] D.K. Chaudhary, R.L. Dua, “Application of Multi objective Particle Swarm Optimization to Maximize
Coverage and Lifetime of wireless Sensor Network”, International Journal Of Computational
Engineering Research, Vol. 2, Issue 5, pp. 1628-1633, 2012.
[24] S.E. Nezhad, H.J. Kamali, M.E. Moghaddam, “Solving K-Coverage Problem in Wireless Sensor
Networks Using Improved Harmony Search”, International Conference Broadband, on Wireless
Computing, Communication and Applications (BWCCA), Fukuoka, pp. 49-55,IEEE, 2010.
[25] A. Dhawan, S.K. Prasad, “A Distributed Algorithmic Framework for Coverage Problems in Wireless
Sensor Networks”, International Symposium on Parallel and Distributed Processing (IPDPS), Miami,
FL, pp. 1-8, IEEE, 2008.
[26] G. Fan, F. Liang, S. Jin, “An Efficient Approach for Point Coverage Problem of Sensor Network”,
International Symposium on Electronic Commerce and Security, Guangzhou City, pp. 124-128, IEEE,
2008.
[27] Z. Bin, M. Jianlin, L. Haiping, “A Hybrid Algorithm for Sensing Coverage Problem in Wireless
Sensor Networks”, International Conference on Cyber Technology in Automation, Control, and
Intelligent Systems (CYBER), Kunming, pp. 162-165, IEEE, 2011.
[28] M. Cardei, M. Thai, Y. Li, W. Wu, “Energy-Efficient Target Coverage in Wireless Sensor Networks”,
In Proceedings of the 24th Annual Joint Conference IEEE Computer and Communication Society
(INFOCOM), Miami, USA, Vol. 3, pp. 1976-1984, IEEE, March 2005.
[29] H. Tan, Y. Wang, X. Hao, Q. S. Hua, F. Lau, “Arbitrary Obstacles Constrained Full Coverage in
Wireless Sensor Networks”, In: WASA, Vol. 6221, pp. 1-10, 2010.
[30] F.P. Quintao, F.G. Nakamura, G.R. Mateus, “Evolutionary Algorithm for the Dynamic Coverage
Problem Applied to Wireless Sensor Networks Design”, Congress on Evolutionary Computation, Vol.
2, pp. 1589-1596, IEEE, 2005.
[31] W.H. Liao, Y. Kao, R.T. Wu, “Ant Colony Optimization based Sensor Deployment Protocol for
Wireless Sensor Networks”, Expert Systems with Applications, Vol. 38, pp. 6599-6605, Elsevier Ltd,
2010.
[32] M.A. Jamali, N. Bakhshivand, M. Easmaeilpour, D. Salami, “An Energy-Efficient Algorithm for
Connected Target Coverage Problem in Wireless Sensor Networks”, 3rd International Conference on
Computer Science and Information Technology (ICCSIT), Chengdu, Vol. 9, pp. 249-254, IEEE, 2010.
[33] X. Wang, G. Xing, Y. Zhang, C. Lu, R. Pless, C. Gill, “Integrated Coverage and Connectivity
Configuration in Wireless Sensor Networks”, In Proceedings of the 1st international Conference on
Embedded Networked Sensor Systems, Los Angeles, California, USA, pp.28-39, 2003.
[34] B. Liu, D. Towsley, “A Study on the Coverage of Large-Scale Sensor Networks”, In The 1st IEEE
International Conference on Mobile Ad-hoc and Sensor Systems, 2004.
[35] J. Kennedy, R.C. Eberhart, “Particle Swarm Optimization”, In Proceedings of the IEEE International
Conference on Neural Networks, pp. 1942-1948, 1995.
[36] Y. Shi, R.C. Eberhart, “A Modified Particle Swarm Optimizer”, In Proceedings of the IEEE
International Conference on Evolutionary Computation, pp. 69-73, Anchorage, AK, 1998.
[37] Y. Shi, R.C. Eberhart, “Parameter Selection in Particle Swarm Optimization”, In Proceedings of the
5rd Evolutionary Computation (EP98), Springer-Verlag, pp. 591-600, San Diego, California, USA,
1998.
[38] R. Storn, K. Price, “Minimizing the Real Functions of the ICEC’96 Contest by Differential Evolution”,
International Conference on Evolutionary Computation, Nagoya, Japan, 1995.
[39] J. Holland, “Adaptation in Natural and Artificial Systems”, University of Michigan, Michigan, USA,
1975.
[40] W.R. Heinzelman, A. Chandrakasan, H. Balakrishnan, “Energy-Efficient Communication Protocols
for Wireless Microsensor Networks”, In Proceedings of the 33rd Hawaii International Conference on
System Sciences, pp. 1-10, IEEE, 2000.
74


Authors
Isa Maleki is a Lecturer and Member of The Research Committee of The Department of
Computer Engineering, Dehdasht Branch, Islamic Azad University, Dehdasht, Iran. He Also
Has Research Collaboration with Dehdasht Universities Research Association NGO. He is a
Member of Review Board in Several National Conferences. His Interested Research Areas Are
in the Software Cost Estimation, Machine Learning, Data Mining, Optimization and Artificial
Intelligence.
Seyyed Reza Khaze is a Lecturer and Member of the Research Committee of the Department
of Computer Engineering, Dehdasht Branch, Islamic Azad University, Dehdasht, Iran. He is a
Member of Editorial Board and Review Board in Several International Journals and National
Conferences. His Interested Research Areas Are in the Software Cost Estimation, Machine
Learning, Data Mining, Optimization and Artificial Intelligence.
Marjan Mahmoodi Tabrizi is a M.Sc. Student in Department of Computer Engineering,
Science and Research Branch, Islamic Azad University, West Azerbaijan, Iran. Her Interested
Research Areas Are in the Wireless Sensor Networks, Data Mining, Optimization and
Machine Learning.
Ali Bagherinia is a Lecturer and Member of the Research Committee of the Department of
Computer Engineering, Dehdasht Branch, Islamic Azad University, Dehdasht, Iran. He Has a
Currently Ph.D Candidate In Department Of Computer Engineering At Science And Research
Branch, Islamic Azad University, Iran. His Interested Research Areas Are in the Wireless
Sensor Networks, Data Mining, Optimization and Artificial Intelligence.

75

A new approach for area coverage problem in wireless sensor networks with hybrid particle swarm optimization and differential evolution algorithms

Recommended

More Related Content

What's hot (16)

Viewers also liked (7)

Similar to A new approach for area coverage problem in wireless sensor networks with hybrid particle swarm optimization and differential evolution algorithms (20)

Recently uploaded (20)

A new approach for area coverage problem in wireless sensor networks with hybrid particle swarm optimization and differential evolution algorithms