Distributed Hybrid Particle/FIR Filtering for Mitigating NLOS Effects in TOA-Based Localization Using Wireless Sensor Networks

Abstract

For indoor localization based on wireless sensor networks, the transmission of wireless signals can be disrupted by obstacles and walls. This situation, called non-line-of-sight (NLOS), degrades localization accuracy and may lead to localization failures. This paper proposes a new NLOS identification algorithm based on distributed filtering to mitigate NLOS effects, including localization failures. Rather than processing all measurements via a single filter, the proposed algorithm distributes the measurements among several local filters. Using distributed filtering and data association techniques, abnormal measurements due to NLOS are identified, and negative effects can be prevented. To address cases of localization failures due to NLOS, the hybrid particle finite impulse response filter (HPFF) was adopted. The resulting distributed HPFF can self-recover by detecting failures and resetting the algorithm. Extensive simulations of indoor localization using time of arrival measurements were performed for various NLOS situations to demonstrate the effectiveness of the proposed algorithm.