Robust PCA based extended target estimation with interference mitigation

A novel approach based on robust principal component analysis (PCA) is proposed in this paper to perform extended target estimation with interference mitigation and learning. Robust PCA can accurately recover the low rank matrix and the sparse matrix from their summation. The data from the estimated target constitutes the low rank matrix while the interference signal contributes to the sparse matrix. From the preliminary results, even with arbitrarily large interference signal, the impulse response or the transfer function of the extended target can be estimated. Thus, the proposed approach can be widely used for anti-interference task in the radar society.