基于改进不确定集的稳健波束形成算法

基于不确定集约束的稳健MVDR波束形成算法在一定程度上依赖于期望信号导向矢量误差的先验知识，且当导向矢量失配较严重时干扰抑制性能也有所下降。为此，提出了一种基于投影变换的改进算法。该方法将约束方向矢量向信号干扰子空间投影，并作为新的约束方向矢量，从而等效于减小了期望信号导向矢量误差。这样，误差不确定参数只需设置为一较小的实数即可在任意导向矢量失配时获得最优的输出性能。计算机仿真结果证明了所提波束形成器具有较强的稳健性能。

A Robust Beamforming Algorithm Based on Modified Uncertainty Set

Since robust MVDR beamforming method based on the constraint of uncertainty set depends on the priori knowledge of the steering vector mismatch of the signal of interest（SOI） in a way and the performance of interference suppression would degrade when the steering vector mismatch is serious, an improved robust MVDR beamforming method is presented in this paper. The constrained steering vector is projected onto the signal-interference subspace as the new one, which is equivalent to decreasing the steering vector error of the SOL Thus, a relatively small real number is sufficient for the error uncertainty parameter, which can achieve the best performance with arbitrary steering vector mismatch. The simulation results demonstrate the robustness of the proposed beamformer.