Multi-Target Delay and Doppler Estimation in Bistatic Passive Radar Systems

We propose a computationally efficient algorithm for localizing multiple targets on a delay-Doppler plane in a bistatic passive radar system. Our algorithm aims to jointly compute the maximum likelihood (ML) estimate of the multiple targets’ (MT) delays and Doppler shifts by developing the expectation maximization (EM) algorithm. The proposed method, referred to as EM-MT, decomposes the multi-dimensional estimation for delays and Doppler shifts into several independent, per-target optimization problems. At the output of EM-MT, we obtain the ML estimate of the delays and Doppler shifts of the multiple targets, and also the estimate of each target’s component signal in the surveillance channel (SC). The latter signal can be used to subtract stronger targets from the SC in order to accurately estimate the weaker targets. Simulation results show that EM-MT outperforms the cross-correlation estimator, the modified CC (MCC) estimator and a previously proposed EM-based algorithm. Furthermore, unlike these algorithms, our proposed EM-MT algorithm achieves the Cramer-Rao lower bound.