基于加权l_1-SRACV算法的稀疏DOA估计

Sparse DOA Estimation Based on Weighted l_1-SRACV Algorithm

针对稀疏重构到达角估计方法中的l1-SRACV算法在低信噪比条件下,估计得到的空间谱伪峰较多的问题,利用子空间的方法选取权值,对l1-SRACV算法的目标函数进行加权,以达到抑制伪峰的目的。阐述了利用子空间方法选取权值的合理性,并讨论了噪声子空间欠估计与过估计时对权值性质的影响。理论分析和仿真实验表明:低信噪比时,加权l1-SRACV算法能够有效地抑制伪峰,并减小了角度估计误差;噪声子空间的欠估计与过估计均会减弱伪峰抑制效果。

In the condition of low signal-to-noise ratio（ SNR）,there are spurious peaks in the spatial spectrum obtained by l1-SRACV algorithm for direction of arrival（ DOA） estimation which is based on sparse reconstruction. In order to suppress spurious peaks,the target function of l1-SRACV algorithm is weighted and the weights are worked out by subspace method in the paper. The rationality of working out weights by subspace method is elaborated and influence to the property of weights is discussed when noise subspace is underestimated or overestimated. Theoretical analysis and experiments show that spurious peaks can be suppressed efficiently and the root mean square error of DOA estimation can be reduced by weighted l1-SRACV algorithm when SNR is low. Suppression to spurious peaks is weakened when noise subspace is underestimated or overestimated.