中文摘要：

为了将盲信号分离应用于波达方向估计，在基于四阶累积量的定点迭代快速独立分量分析（Independent Component Analysis，ICA）算法进行盲信号分离的基础上，利用分离矩阵得出混合矩阵的估计，并对混合矩阵的列向量在真实阵列流型上进行投影，通过角度扫描估计出信号的方位角。仿真结果表明，该算法在信噪比较高的条件下，具有跟MUSIC（Multiple Signal Classification Method）算法相似的分辨性能，但是在信噪比较低的情况下表现出较高的分辨率。

英文摘要：

To apply the blind signal separation method to DOA estimation, the blind signal separation （BSS） algorithm based on fast fixed - point independent component analysis is introduced in this paper. By using this algorithm, the estimation of manifold matrix is deduced from the gained separated matrix. We obtain the DOA （direction - of - arrival） of signals from angle scaling based on the projection of the column vector on the real array configuration. The results of the experiment show that this algorithm has good resolution in a low SNR （ Signal Noise Ratio） environment compared with the MUSIC （Multiple Signal Classification） algorithm.