A Descend-Based Evolutionary Approach to Enhance Position Estimation in Wireless Sensor Networks

Wireless sensor networks have wide applicability to many important applications including environmental monitoring and military applications. Typically with the absolute positions of only a small portion of sensors predetermined, localization works for the precise estimation of the remaining sensor positions on which most location sensitive applications rely. Intrinsically, localization can be formulated as an unconstrained optimization problem based on various distance/path measures, for which most of the existing work focus on increasing its precision through different heuristic or mathematical techniques. In this paper, we propose to adapt an evolutionary approach, namely a micro-genetic algorithm (MGA), and its variant as postoptimizers to enhance the precision of existing localization methods including the Ad-hoc Positioning System. Our adapted MGA and its variants can easily be integrated into different localization methods. Besides, the prototypes of our evolutionary approach gained remarkable results on both uniform and anisotropic topologies of the simulation tests, thus prompting for many interesting directions for future investigation