自适应波束形成算法性能很大程度上依赖期望信号到达角信息。当期望信号导向矢量存在失配,期望信号将会被抑制,阵列输出信干噪比会下降。传统空间投影算法将导向矢量向信号子空间投影,能够改善一般导向矢量失配的稳健性,但是子空间维数很难确定。文中提出的算法利用协方差矩阵自身的特性,以及线阵阵列流形矢量对称特点,对协方差矩阵取前后向平均,利用协方差矩阵逆和最小特征值比值的高阶次幂估计信号子空间,可有效解决子空间维数不确定的问题,并能在较少快拍次数下达到抗干扰目的。计算机仿真结果证明新算法的正确性和有效性。 The performance of the adaptive beamforming algorithm relies heavily on the desired signal arrival angle information. When there is a mismatch between the signal steering vectors, the expected signal will be suppressed and the array output SINR will decrease. The traditional space projection algorithm projects the steering vector to the signal subspace, which can improve the robustness of the general steering vector mismatch, but the dimension of subspace is hard to be determined. The proposed algorithm utilizes the covariance matrix’s own characteristics and the vector symmetry of the manifold array manifold to take the forward and backward averaging of the covariance matrix and estimates the signal subspace using the higher order power of the covariance matrix inverse and the minimum eigenvalue ratio , Which can effectively solve the problem of uncertain dimensionality in subspaces and achieve the purpose of anti-jamming with fewer snapshots. Computer simulation results show that the new algorithm is correct and effective.