High-Accuracy DOA Estimation based on Vectorized Fourth-Order Cumulant with Coprime Array

Abstract

The existing fourth-order cumulant direction of arrival (DOA) estimation methods are mainly based on the uniform linear array, which have high degree of freedom, but poor estimation accuracy. In order to improve the DOA estimation accuracy of the fourth-order cumulant methods, a vectorized fourth-order cumulant DOA estimation method based on coprime array (CA-VFOC) is proposed. Firstly, the fourth-order cumulant matrix of the received signals is computed. Then, according to the structure characteristics of sparse coprime array, the fourorder cumulant matrix is vectorized to construct the virtual array model, and the continuous response part of the virtual array model is extracted. Finally, the DOA estimation cost function is constructed by spatial smoothing, and the complexity analysis of the algorithm is given. Simulation results show that compared with the fourth-order cumulant method of uniform array and the traditional fourth-order cumulant method of coprime array, the proposed method can effectively improve the DOA estimation accuracy, and can achieve the effective DOA estimation of multitarget signals in the underdetermined conditions with the same physical sensor number.