Dynamic Deployment of Sensors using VirtualForces-directed Co-Evolutionary Particle SwarmOptimization (VFCPSO) Method

Abstract—
Wireless sensor networks (WSN) present
numerous research possibilities due to the bright
and promising future of information technology.
WSNs are a group of inexpensive, less power
consuming, having multiple functions and tiny
wireless nodes that work in unison. They sense the
environment; do simple tasks like processing of
data and are able to communicate wirelessly over
a short distance. How effective a WSN will depend
on the probability of coverage and on detecting
target. Many coverage strategies like force
based (Virtual Forces Algorithm), Grid-Based
(triangular lattice, hexagon and square grid)
and computational geometry based (Voronoi and
Delaunay triangulation) are suggested. Another
method is the Particle Swarm Optimization
method (PSO). This is a type of Evolutionary
Algorithm and shows prospects in being able
to solve complex optimization problems. The
demand to achieve optimum coverage inspires
us to move towards hybrid algorithms. The
hybrid algorithms combine more than one of
the above mentioned approaches. We aspire to
achieve optimum Coverage by implementing
an algorithm called Virtual Force-directed
Co-evolutionary Particle Swarm Optimization
(VFCPSO). This algorithm is a hybrid of Virtual
Forces Algorithm and Co-evolutionary PSO.
It is flexible enough for a network formed from
permutations of homogeneous, heterogeneous,
stationary and mobile sensors. The VFCPSO
predicts deploying actively with better abilities to
search overall and to converge reginally. We aim
to implement and compare the results of VFA,
PSO, VFPSO and VFCPSO. The expectation
here is to get a noticeable increase in effective
coverage area and a noticeable decrease in
average computation time.