Cooperative localization using angle of arrival measurements: sequential algorithms and non-line-of-sight suppression

摘要

We investigate localization of a source based on angle of arrival (AoA)measurements made at a geographically dispersed network of cooperatingreceivers. The goal is to efficiently compute accurate estimates despiteoutliers in the AoA measurements due to multipath reflections innon-line-of-sight (NLOS) environments. Maximal likelihood (ML) locationestimation in such a setting requires exhaustive testing of estimates from allpossible subsets of "good" measurements, which has exponential complexity inthe number of measurements. We provide a randomized algorithm that approachesML performance with linear complexity in the number of measurements. Thebuilding block for this algorithm is a low-complexity sequential algorithm forupdating the source location estimates under line-of-sight (LOS) environments.Our Bayesian framework can exploit the ability to resolve multiple paths inwideband systems to provide significant performance gains over narrowbandsystems in NLOS environments, and easily extends to accommodate additionalinformation such as range measurements and prior information about location.