Genetic algorithm for solving minimal exposure path in mobile sensor networks

Abstract

The Minimal Exposure Path (MEP) is a fundamental factor to measure quality of service in Wireless Sensor Networks (WSNs). The MEP not only improves the performance of WSNs but also guides the action of agents in the sensing field. Up to now, MEP has been focusing on solving problems in stationary sensor networks. MEP with both high dimensionality and non-linear property in stationary WSNs is NP-hard problem. This paper formulated a problem of finding the MEP in Mobile Sensor Networks (MSNs), in which sensors' mobility plays an important role in the execution of application, and then proposed a genetic algorithm to solve it. Our methodology is not only to calculate upper and lower bounds of exposure value, but also to determine the MEP trajectory. The proposed algorithm was experimented on 30 random instances with various numbers of mobile sensors and different velocities of intruder which are proportional to the velocity of sensors. We also compared the minimal exposure value obtained from MSNs model with the one obtained from static sensor network model. Experimental results showed that our genetic algorithm is stable, convergent and effective when applying in both mobile and static models of WSNs.