Two-Phase Stochastic Optimization to Sensor Network Localization

Abstract

In this paper we describe a novel approach to sensor network localization, i.e., two-phase algorithms based on simulated annealing and genetic algorithm. The numerical results presented and discussed in the final part of the paper show that these novel schemes give accurate and consistent location estimates of the nodes in the network. The performance is better and speed is faster than that of the semidefinite programming (SDP) and one-phase simulated annealing (SA).