Improvement of 2-D direction finding algorithm based on two L-shape arrays

This paper presents an improved method of two-dimensional (2-D) direction finding based on two L-shape arrays. A new type of three-dimensional (3-D) direction angles has been defined in this method. The angles along each coordinate axis estimated by the rotation invariant algorithm (ESPRIT) of uniform linear array (ULA). All estimated angles got pair-matched by spatial relationships of their direction cosines. For resolving the problem which ULA is sensitive to direction-of-arrival (DOA), weighted least-square (WLS) method is used to refine the estimated angles. Simulation results show that the proposed method, compared with the conventional methods, has higher estimation accuracy and is not sensitive to DOAs.