Multi-layer Genetic Algorithm for Maximum Disjoint Reliable Set Covers Problem in Wireless Sensor Networks

Abstract

Densely deployment of sensors is generally employed in wireless sensor networks to ensure complete target coverage for a long period of time. Many sensors scheduling techniques have been recently proposed for prolonging the network lifetime. Scheduling sensors into a maximum number of disjoint sets has been modeled, in the literature, as disjoint set covers (DSC) problem which is a well-known NP-hard optimization problem. Unlike other attempts which considers only a simple disk sensing model, this paper addresses the problem of finding the maximum number of set covers while considering a more realistic sensing model to handle uncertainty into the sensors' target-coverage reliability. The paper investigates the development of a simple multi-layer genetic algorithm (GA) whose main ingredient is to support selection of a minimum number of sensors to be assigned to maximum number of set covers. With the aid of the remaining unassigned sensors, the reliability of DSCs provided by the GA, can further be enhanced by a post-heuristic step. By identifying the upper bound of number of disjoint set covers, the GA can successively construct covers, each of which is of minimal sensor cost. Performance evaluations on solution quality in terms of both number of set covers and coverage reliability are measured and compared through extensive simulations, showing the effectiveness of the proposed GA.