中文摘要：

现有基于多级维纳滤波（MSWF）的子空间法波达方向（DOA）估计算法复杂度较低,但需要先验的参考信号.论文从MSWF求解线性预测问题入手,将基于MSWF的线性预测和子空间两种不同的DOA估计方法结合起来,提出了一种实用的低复杂度DOA估计算法.该算法无需构造专门的参考信号,在低信噪比或信源数估计不准的情况下,算法依然具有较好的稳健性和估计性能.仿真实验验证了本文的结论.

英文摘要：

The current subspace algorithm for direction of arrival（DOA） estimation based on multi-stage wiener filter（MSWF） has low computational complexity,but it relies heavily on the prior knowledge of a reference signal,which is often unavailable.In this paper,starting from solving a linear prediction（LP） problem using MSWF,we propose a new DOA estimation algorithm with low computational complexity.The new algorithm combines the LP method with the subspace method,and does not require the reference signal.Further,the new algorithm is shown to be stable and has a good estimation performance,even in the case of low signal to noise ratio（SNR） or incorrect source number estimation.Simulation results are provided to verify our conclusion.