Using Micro-Genetic Algorithms to Improve Localization in Wireless Sensor Networks

Abstract

Wireless sensor networks are widely adopted inmany location-sensitive applications including disastermanagement, environmental monitoring, militaryapplications where the precise estimation of each nodeposition is inevitably important when the absolute positionsof a relatively small portion as anchor nodes of theunderlying network were predetermined. Intrinsically,localization is an unconstrained optimization problem basedon various distance/path measures. Most of the existinglocalization methods focus on using different heuristic-basedor mathematical techniques to increase the precision inposition estimation. However, there were recent studiesshowing that nature-inspired algorithms like the ant-basedor genetic algorithms can effectively solve many complexoptimization problems. In this paper, we propose to adaptan evolutionary approach, namely a micro-geneticalgorithm, as a post-optimizer into some existing localizationmethods such as the Ad-hoc Positioning System (APS) tofurther improve the accuracy of their position estimation.Obviously, our proposed MGA is highly adaptable andeasily integrated into other localization methods.Furthermore, the remarkable improvements attained byour proposed MGA on both isotropic and anisotropictopologies of our simulation tests prompt for severalinteresting directions for further investigation.