Design, Development and Experimental Validation of Multi-Target Tracking Framework for Passive Radar

In this paper we present the multi-target tracking (MTT) framework for passive radar (PR). We address the problem of target filtering in the presence of clutter and measurement uncertainty by employing a two-stage cascaded logic for track initiation. We discuss target association in both plot-to-plot and plot-to-track association regimes, and give practical solutions for efficient data association in multi-target environment. We demonstrate the effectiveness of proposed framework using Kalman filter (KF) based range-doppler (RD) tracking of multiple targets, and verify the system performance against truth data in both bi-static and cartesian domain. Radar measurements have been taken using a mobile PR system, developed at Prince Sultan Defense Studies and Research Center (PSDSARC), Saudi Arabia. Experiments conducted with arbitrary transmitter-target-receiver geometries, exploiting a single non-cooperative FM broadcast channel in bi-static configuration, show that the radar is able to reliably track aerial targets at bi-static ranges of beyond 300 Km, corresponding to ranges up to 150 Km from the receiver.