基于特征值分解的稳健波束形成技术

Capon 自适应波束形成中，导向矢量误差和协方差矩阵的估计误差均会导致波束形成器的性能下降。针对这一现象，提出一种基于特征值分解的稳健波束形成技术，即在将估计的协方差矩阵特征值分解后，直接对影响波束形成器稳健性能的噪声小特征值进行算术平均处理，以获得接近于理想波束形成器的稳健性能。同时根据试验仿真，给出了用于区分干扰和噪声特征值的门限计算公式，为准确构建特征子空间提供了思路。分析结果表明，与传统的对角加载方法相比，该方法能够达到同样的改善性能，在实际运用中更加直接和有效。

Robust Beamforming Technique Based on Eigenvalue Decomposition

Regarding the problem of the performance degradation of Capon adaptive beamformer due to steering vector error and the estimation error of covariance matrix, this paper proposes a new method that by way of averaging the noise eigenvalue which mainly effect on the robustness of the beamformer after the eigenvalue decomposition. An estimation of threshold value is obtained from the results of simulation experiments; and compared with traditional diagonal loading methods, this technique can achieve the same performance and be more effective and direct in practical application.