非均匀噪声环境下基于矩阵补全的DOA估计方法

Direction-of-Arrival Estimation Method Based on Matrix Completion in Presence of Non-Uniform Noise

针对传统的信号波达方向(DOA)估计算法无法适用于实际应用中非均匀噪声、数据不完整等情况的问题,提出了一种结合矩阵补全理论和最大似然交替投影算法的DOA估计方法。在背景噪声为非均匀噪声的情况下,该方法通过对只有部分元素已知的阵列协方差矩阵进行矩阵补全,将稀疏矩阵重构为无噪声协方差矩阵,然后利用最大似然交替投影算法实现对DOA的估计。实验仿真表明:该DOA估计方法能够有效恢复不完整数据并抑制非均匀噪声的影响,而且在低信噪比条件下,仍具有较好的DOA估计性能。

The traditional direction-of-arrival(DOA)estimation algorithm can not be directly utilized when the noise is non-uniform or the data are incomplete in practical applications.Thus,we proposed a DOA estimation method combining matrix completion theory and maximum likelihood alternating projection algorithm.When the spatially non-uniform noise was presented,the matrix completion method was first utilized to reconstruct a noiseless covariance matrix with the use of sparse matrix.Then the maximum likelihood alternating projection algorithm was applied to estimate DOA.Numerical simulations show that the proposed DOA estimation method can recover incomplete data effectively and suppress the influence of non-uniform noise.Additionally,the proposed method still has better DOA estimation performance than the other methods in low signal-to-noise ratio scenarios.