两种解相关算法的DOA估计性能研究

Study of DOA Estimation Performance for Correlated Signals by Two Decorrelating Algorithms

MUSIC算法能对独立信号的波达方向（DOA）进行有效估计，但该算法对相干信号或强相关信号进行DOA估计时，算法性能会随着相关系数的增加而急剧恶化甚至失效．针对该问题，利用前后向空间平滑技术与修正MU—SIC算法进行去相关，研究了它们去相关的基本原理及优缺点；利用MUSIC算法通过实验仿真对相关信号的波达方向估计性能进行了研究，分析了信号间相关系数对前后向空间平滑算法与修正MUSIC算法的测向性能的影响，并验证了算法去相关的有效性．

MUSIC algorithms can process DOA estimation for uncorrelated signals. This algorithm gives poor performance or even is disabled in coherent signals or strong correlation. To solve this problem, the paper studies the decorrelating algorithms, forward-back spatial smoothing and mod- ified MUSIC. In the paper, the correlated signals＇ DOA estimation performance is analyzed in de- tail by using the two algorithms, and the algorithms~ decorrelating theory and characteristics are also studied. Finally,the simulation results about the correlated coefficient and algorithms＇ DOA performance under different correlated coefficients is given, and the paper analyses the interrela- tion between correlated coefficient and algorithms＇ DOA performance, and show that the algo- rithms are effective in decorrelation.