Direction-of-arrival estimation of coherent and noncoherent mixed signals under unknown mutual coupling

The performance of high resolution direction-of-arrival (DOA) estimation based on the orthogonality of subspaces is degraded by array manifold perturbation resulted from mutual coupling (MC) and rank loss of covariance matrix caused by the coherency of the signal. In this paper, a direction finding method, without any compensation, was proposed for coherent and non-coherent mixed signals in the presence of unknown MC. The method automatically suppressed the effect of unknown MC by calculating the covariance matrix with part of the received data. Then, an equivalent covariance matrix was reconstructed with the eigenvectors gained by eigen-decomposition of the covariance matrix. By performing the singular value decomposition (SVD) of the equivalent covariance matrix, the DOA of mixed signals could be estimated via MUSIC or ESPRIT algorithms. The proposed algorithm improved the conventional SVD method in order to estimating the DOA of coherent and noncoherent mixed signals. Without any array calibration, the algorithm achieves a favourable accuracy of DOA estimation under unknown MC. The simulation results demonstrate the validity of the proposed method. ©, 2015, Shanghai Jiao Tong University. All right reserved.