Abstract: A fast algorithm for direction-of-arrival (DOA) estimation is presented. By the characteristic that the mutual correlation vector of a uniform array is located in the signal subspace, a signal subspace estimation method is obtained using a multi-stage wiener filter, which avoids the known expected signal and eigen-decomposition. Then, the min-norm rooting method is adopted to realize DOA estimation. The method can estimate DOA effectively even in the case of no expected signal, and the complexity is O(NLP), where N is the number of sensors, L is the number of training sets, and P is the number of signals. Finally, simulation results are given to demonstrate the effectiveness of the method.

Key words: DOA estimation, multi-stage wiener filter, min-norm solution, low complexity