A Geometric Filter Algorithm for Robust Device-Free Localization in Wireless Networks

Abstract

Device-free localization (DFL) is the method of determining an uncooperative target’s location by using the shadowing induced by the target on the radio links crossing through the wireless network’s deployment area. Most DFL schemes require high memory and computing resources for accurate tracking performance, and thus may not be suitable for resource-constrained applications. To address this issue, a geometric filter (GF) algorithm that achieves accurate tracking using only operations on simple geometric objects is proposed. It reduces the algorithm’s storage and computational resource requirements resulting in fast execution times. Experimental results show that the GF algorithm outperforms some grid-based and particle filter-based DFL methods with up to $65\%$ reduction in tracking root-mean-square error, and up to 300 times improvement in average execution time.