Two-Dimensional Direction of Arrival Estimation Using Generalized ESPRIT Algorithm with Non-uniform L-Shaped Array

The aim of this study is to solve the problem of two-dimensional direction of arrival (2D-DOA) estimation for non-uniform L-shaped array by employing generalized ESPRIT (GESPRIT) algorithm. GESPRIT algorithm can be seen as an extension of conventional ESPRIT algorithm, which doesn't require any particular array geometry. Our work is to extend this method to the 2D-DOA estimation case. The 2D-GESPRIT algorithm, which is referred to as the direct extension of GESPRIT algorithm, performs poorly in actual implementation. We make improvement by exploiting the 2D-NGESPRIT algorithm with reference to the NGESPRIT algorithm. To reduce the computational complexity, we propose a successive GESPRIT (S-GESPRIT) algorithm which needs only one-dimensional searches. The S-GESPRIT algorithm has comparable performance to the 2D-NGESPRIT algorithm and needs no additional pair-matching procedure. Furthermore, it imposes no constraints on the sensor spacing. We also derive the Cramer-Rao bound of 2D-DOA estimation for a non-uniform L-shaped array and conduct the computational complexity analysis. The simulation results verify the effectiveness and improvement of the proposed algorithms.