Range-Doppler processing for passive radar based on joint tracking and dynamic compressed sensing

Passive radar (PR) systems use accessible sources of illumination for tasks such as detection and tracking. Improving the performance of range-Doppler (RD) processing is essential to subsequent detection of PR systems. Traditional RD processing methods such as cross-ambiguity function (CAF) and compressed sensing (CS) have limited performance when dealing with weak targets. In this paper, we propose a RD processing method in PR based on joint tracking and dynamic compressed sensing (DCS), which can more accurately reconstruct RD maps by utilizing inter-frame motion information of the target. The cubature Kalman filter (CKF) is employed to predict the target motion state of each RD map in the proposed method. Then a dynamic extended orthogonal matching pursuit (DEOMP) is presented to generate more accurate RD maps by utilizing the prediction information to determine the support set. Numerical simulations are conducted to validate the performance of the proposed method.