A Single-Mobile-Anchor Based Distributed Localization Scheme for Sensor Networks

This paper investigates the distributed localization problem for a sensor network in the plane. It is assumed that one mobile robot wandering in the area acts as a mobile anchor. A distributed scheme is devised such that each node can localize itself using only its locally available distance measurements about its (possibly changing) neighbors. The key idea is based on a new way of calculating the barycentric coordinate of each node about its neighbors, which has the advantage that it is capable of handling distance measurement errors. A necessary and sufficient condition is obtained for the global convergence of the proposed distributed algorithm. Both simulation and experiment using ZigBee devices are carried out to validate the proposed distributed localization scheme.