Coverage-Enhancing algorithm for video sensor network based on
improved particle swarm optimization
Xiang Fu
1, a
, Jiexian Zeng
2, b
1
School of Software, Nanchang Hangkong University, Nanchang, 330063, China
2
School of Software, Nanchang Hangkong University, Nanchang, 330063, China
a
email: fxfb163@163.com,
b
email: zengjx58@163.com
Keywords: Video sensor networks; Sensing coverage; Particle swarm optimization (PSO)
Abstract. The parameters of traditional particle swarm optimization (PSO) methods are
unchangeable, which may lead to the iterations have slow convergence speed or unable to converge
to global optimum. In this paper, improved PSO algorithm is applied to the coverage of video
sensor network. Through improving inertia factors of PSO, the algorithm has great local search
ability thus can avoid converging to local optimum value, and it can converge quickly to the global
optimum value. Experiment results show that the proposed method has faster convergence speed
and better coverage rate than traditional method based on PSO.
Introduction
Wireless sensor networks can be deployed quickly, have the ability of self-organization and are
high concealment, so they are increasingly popular in a number of application domains such as
environment supervision, traffic control and battlefield reconnaissance. In order to supply visual
image or video data, video sensor network is widely applied because of the technology development
and the cost reduction. Coverage enhancement is a fundamental problem in wireless sensor
networks, and many researchers have carried out researches in this area [1,2,3]. Most of related
researches are based on an omni-directional sensing model. However, video sensor can sense the
area within a limited angle only, which is a directional sensor, so traditional research results about
the coverage of wireless sensor networks can not be used into wireless video network directly. Thus,
coverage enhancement for directional sensor networks has drawn more attention and has become a
focus.
Ma et al [4] proposed the concept of model for directional sensing firstly, and explored the
integrity and connectivity problems of directional sensing networks. Potential field-based
directional sensing coverage enhancing algorithms [5,6,7] have been made a lot of researches. In
this kind of methods, each camera is regarded as a virtual particle and can be repelled by neighbor
cameras, thus the problem of nodes’ direction adjustment is converted into the problem of particles’
union distribution, and then the coverage of the directional network can be enhanced effectively
through decreasing overlapping and blind area. However, when the forces between nodes are in
equilibrium, the directions of nodes are not necessarily the most optimum [8]. But because the
directions cannot move anymore, sometimes the coverage rate is affected. In order to overcome the
shortage of potential field-based methods, Xu et al [8][9] proposed to solve the coverage of video
sensor network by using PSO. Since traditional PSO is easy to fall into local optima [10], the
mentioned coverage methods of video sensor network based on PSO have similar disadvantages
such as local optima and low convergence speed.
In this paper, improved PSO algorithm is used in the coverage enhancing of video sensor
network to overcome the shortages of traditional methods based on PSO. Experiments show that the
proposed method has higher convergence speed and better coverage performance than traditional
coverage method based on PSO.
International Conference on Intelligent Systems Research and Mechatronics Engineering (ISRME 2015)
© 2015. The authors - Published by Atlantis Press