Robust Localization for Wireless Sensor Networks

Moore at al. presented in [1] a cluster-based distributed algorithm for localizing wireless sensor network (WSN) nodes in presence of range measurement noise. They introduced the notion of "robust quadrilaterals" as a means of avoiding flip ambiguities that often corrupt the results of localization algorithm. This paper¹ proposes an enhancement of the algorithm presented in [1] that uses a new quadrilateral robustness test. A second novel distributed algorithm for anchor-less localization of WSN nodes in presence of range measurements noise is presented. This second algorithm reduces the total computational complexity with respect to the first two algorithms by a factor at least equal to the number of nodes in the WSN. It is made of two main steps: the first one, where a single cluster of nodes is localized, and a second one where the other unknown nodes are estimated by successive trilaterations. The original algorithm [1] and the two novel algorithms proposed in this paper are compared by means of simulations in terms of complexity, convergence and location accuracy performance.