Guaranteed localization using reliability of measurements in imperfect mobile sensor networks

This paper deals with localization problems in mobile sensor networks. Such problems are challenging especially when imperfections may occur. In this paper, we propose a robust localization technique that works efficiently under imperfect circumstances. The proposed method is an anchor-based method where two types of sensors are considered: anchors, having known positions, and non-anchor nodes, unaware of their locations, and thus they need to be localized. To do so, nodes collect distance information from neighboring anchors. In addition to anchors information, the reliability of exchanged messages is assumed to be known. Each node combines then all available information to make a final decision about its position. The combination is performed using both the Dempster-Shafer and the interval theories. Using these techniques, nodes are able to compute accurate estimates of their positions even in presence of unreliable measurements. Simulation results corroborate the efficiency of the proposed method, compared to methods developed for perfect environments.