Nonlinear filtering for emission source tracking using biased RSS measurements

Abstract

Under indoor environments, positioning and tracking using GPS and radar measurements are very scarce. Comparatively, positioning and tracking using received signal strength (RSS) measurements from wireless sensor networks are gaining more and more attention. However, so far all localization or tracking algorithms did not take systematic sensor biases into account. If the biases are not corrected, they will lead to degradation in tracking performance. In this paper, we propose a framework to jointly estimate the dynamic source state and static sensor biases using nonlinear filters such as Extended Kalman filter (EKF) and Unscented Kalman Filter (UKF). Numericals examples show that this framework can estimate both source state and sensor biases very well.