An OS-ESPRIT algorithm for estimating DOA in wireless communications

This paper proposes an OS-ESPRIT algorithm based on one step eigenvalue decomposition (EVD) to estimate the directions of arrival (DOAs) of narrowband signals in wireless communication environments, after analyzing the features and performances of DOA estimation algorithms. The proposed algorithm makes use of one step EVD to estimate DOAs. Compared with the previous works, the new approach calls for low computational complexity, but it exhibits superior performance, such as smaller estimation error and better robustness to signal-to-noise ratio (SNR) change etc. It is applicable to a lower SNR scenario. Simulation results show the effectiveness of the method.