运用滑窗定阶的噪声子空间分解

地面接收的导航信号易受人为干扰，用空时二维阵列处理可有效抑制接收信号中的干扰。多级维纳滤波（MWF: Multistage Weiner Filter）可用于空时二维处理以避免大矩阵特征分解，但噪声子空间估计不准。对此本文提出一种方法:首先用经典MWF粗略估计噪声子空间维数，然后运用滑窗逐步找到噪声和白噪声子空间的分界点，最后用MWF的综合部分即可算得最优权值以进行抗干扰处理。仿真证明该方法在噪声子空间估计上比传统MWF方法具有更高的区分度，并且获得更优的抗干扰效果。由此得出结论:该方法能够大幅提高噪声子空间估计的鲁棒性，增强空时二维阵的抗干扰能力。 

Noise Subspace Decomposition Utilizing Sliding Window Rank Estimation

The received navigation signal on the ground is vulnerable to interferences,which can be suppressed by space-time processing. Multistage Weiner Filter (MWF) is used in space-time processing to avoid large matrix decomposition, but the noise subspace estimation of classic MWF is inaccurate. A method is proposed to solve the problem: first, utilizing conventional MWF to roughly estimate the noise subspace dimension; then, using a sliding window to find the boundary between noise and white noise; finally, utilizing the integrated part of MWF to deduce the optimal weight, which is used to suppress interferences. Through simulation, it is proved that the distinguish capacity of the proposed method is superior to the conventional MWF on noise subspace estimation, thus better interference suppression effect is obtained. It can be concluded that: the proposed method is robust in noise subspace estimation, which enhances the interference suppression capacity of space-time array.