A Novel NLOS Mobile Node Localization Method in Wireless Sensor Network

Abstract

Mobile location estimation using wireless sensor network (WSN) is one of the important technologies in the past decades. A main challenge in mobile location is non line of sight (NLOS) propagation caused by the complicated practical environments. The NLOS effects will seriously reduce the accuracy of localization algorithms. In this paper, we present a novel NLOS mobile node localization algorithm based on probabilistic data association. First, we propose an improved residual weighting (Rwgh) algorithm which can efficiently mitigate the NLOS errors to provide different position estimates by using different subgroups of range measurements. And then a probabilistic data association filter is employed to determine the final position estimate. Experimental results demonstrate that the proposed algorithm is robust to the NLOS errors and provides higher location accuracy over the other traditional filter methods.