针对众多性能优良的超分辨DOA估计算法大都是以预知信源数为前提,信源数估计不准可能会导致DOA估计失败这一问题,提出了一种基于协方差矩阵对角加载的超分辨DOA估计算法。该算法不需预判信源个数和进行特征值分解,且通过对协方差矩阵进行对角加载,可以平滑小快拍数时噪声特征值分散程度,因此该算法更适用于快拍数较少的情况。理论分析表明:该算法的统计估计性能接近于MUSIC(Multiple Signal Classifjcation)算法。计算机仿真结果验证了该算法的鲁棒性和可行性。
Many of the super resolution DOA estimation algorithms presume the sources number are known,but if the sources number are inaccurate, the DOA estimation may fail. To solve it, a super resolution DOA estimation algorithm which is based on diagonal loading is proposed. The algorithm is able to skip sources number estimation and EVD (EigenValue Decomposed), furthermore through diagonal loading on the eovari- ance matrix,it can smooth the dispersive degree of the noise's eigenvalues because of the small amount of snap- shots, so it is more suitable for the small amount of snapshots. The academic analysis indicates that the esti- mated performance of our algorithm is approached to MUSIC algorithm. The feasibility and robustness of the algorithm are verified by the computer simulations.