Robust node localization for wireless sensor networks

Abstract

The node localization problem in Wireless Sensor Networks has received considerable attention, driven by the need to obtain a higher location accuracy without incurring a large, per node, cost (dollar cost, power consumption and form factor). Despite the efforts made, no system has emerged as a robust, practical, solution for the node localization problem in realistic, complex, outdoor environments. In this paper, we argue that the existing localization algorithms, individually, work well for single sets of assumptions. These assumptions do not always hold, as in the case of outdoor, complex environments. To solve this problem, we propose a framework that allows the execution of multiple localization schemes. This multi-modality enables robustness against any single protocol failure, due to its assumptions. We present the design of the framework, and show a 50% decrease in localization error in comparison with state of art node localization protocols. We also show that complex, more robust, localization systems can be build from localization schemes that have limitations.