Bio-Inspired Computation for Solving the Optimal Coverage Problem in Wireless Sensor Networks: A Binary Particle Swarm Optimization Approach

Abstract

The optimal coverage problem (OCP) in a wireless sensor network is to activate as few nodes as possible to monitor the area in order to save energy, while at the same time meeting the full coverage surveillance requirement. This chapter formulates the OCP as a 0/1 programming problem and proposes to use a binary particle swarm optimization (BPSO) algorithm to solve the problem. First, the OCP is modeled as a 0/1 programming problem, where 1 means the node is active and 0 means the node is turned off. This model provides a very natural and intuitive way to interpret the representation to the real network. Second, by considering that the bio-inspired computation algorithms have strong global optimization ability and are very suitable for solving the 0/1 programming problem, this chapter proposes to use the BPSO approach to solve the OCP, resulting in an efficient solution to the OCP. Simulations have been conducted to evaluate the performance of the proposed approach. Moreover, a genetic algorithm (GA) approach is adopted for comparison in the experiments in order to demonstrate the advantages of BPSO in solving the OCP problem. The experimental results show that our proposed BPSO approach not only outperforms the state-of-the-art approaches in minimizing the active-node number, but also performs better than the GA approach in solving the OCP problem under different network scales and different network densities. Moreover, the proposed BPSO approach has very good performance in maximizing the disjoint-set number when compared with the traditional heuristic approaches. © 2015 Elsevier Inc. All rights reserved..