High-resolution direction-of-arrival estimation using Khatri-Rao product and spatial sparsity of sources

Abstract

Abstract. Algorithms based on Khatri-Rao (K-R) product improve the degree of freedom (DOF) of virtual arrays and thus more source angles can be estimated. However, the estimation accuracy is not satisfying. In this paper, a method considering spatial sparsity (SS) of sources in K-R subspace is proposed to estimate direction-of-arrival (DOA) angle. The spurious signal of the virtual array is obtained in K-R subspace and then the DOA angle estimation is finally equal to the solution of a convex optimization problem. In the proposed K-R SS method, high estimation accuracy is achieved by taking advantage of the cross-covariance matrix information. Simulation results demonstrate the effectiveness of the proposed method.