Compressive Sensing-Based Multipath Exploitation for Stationary and Moving Indoor Target Localization

Abstract

Compressive sensing (CS)-based multipath exploitation has been successfully applied to stationary indoor scenes in through-the-wall radar imaging (TWRI). The benefits of using significantly reduced data are also desirable for moving targets. Hence, we bring CS based multipath exploitation to the non-stationary target domain and are able to treat moving and stationary targets simultaneously. In general, multipath propagation has adverse effects on the image quality. However, by using proper modeling, multipath can be used to one’s advantage. In this paper, we apply CS to both stationary and moving targets under interior wall scatterings. Assuming knowledge of the room geometry, we develop an effective method that solves the inverse problem of joint localization and velocity estimation of the targets in an indoor multipath environment from a few measurements. We also propose a computationally inexpensive scheme that first locates the targets using sparse reconstruction and subsequently estimates the velocity vectors. Effectiveness of the proposed methods is demonstrated using both simulated and experimental data.