Interference cancellation based on compressive sensing for passive coherent radar (PACOR)

Interference cancellation is crucial for passive radar and has been well developed. A novel interference cancellation model based on compressive sensing (CS) is proposed in this paper. The sparse matrix is constructed by exploiting the sparsity of the interference signal in delay-Doppler domain, with each column of which is the cross ambiguity function (CAF) of a delayed and Doppler shifted replica of the reference signal. However, the generation of the sparse matrix will be computationally extensive by routine processing, so a fast method is developed based on the fact that all the CAFs can be obtained from the self-ambiguity function (SAF) of the reference signal by approximating. The orthogonal matching pursuit (OMP) algorithm is then applied to obtain the estimator of the sparse vector, which includes the information of the complex amplitude, delay and Doppler of the interference signal. With these parameters, the interference signal is reconstructed and removed from the surveillance signal. Besides dealing with both the stationary and moving clutter, the proposed cancellation algorithm can eliminate the strong target if required by modifying parameters accordingly. Simulation results show that the residual power of interference after cancellation is extremely low, which is meaningful for weak target detection.