四元数域干扰加噪声协方差矩阵重构鲁棒自适应波束形成  

Quaternion-valued Robust Adaptive Beamforming with Interference and Noise Covariance Matrix Reconstruction

作　　者：王一[1] 何冰松[1] 

机构地区：[1]北京理工大学信息与电子学院,北京100081

出　　处：《信号处理》2016年第5期618-622,共5页Journal of Signal Processing

摘　　要：基于电磁矢量传感器阵列的四元数Capon波束形成器较传统的复数域Capon波束形成器有更好的性能。但是该方法在存在指向误差和极化失配的情况下性能急剧下降,甚至会出现信号相消现象。本文将协方差矩阵重构方法推广于四元数Capon波束形成中,通过利用Q-Capon的极化-角度谱估计得到干扰和噪声的功率来对干扰加噪声协方差矩阵进行重构,避免了对角加载方法中对对角加载因子的求解,而且能够有效克服指向误差与极化失配带来的性能下降。计算机仿真表明,该方法相较于其他四元数域的方法有着更好的性能。Q-Capon beamformer which based on the electromagnetic vector sensor array of quaternion field compared with the Capon beamformer in the traditional complex field has a better performance. However,this method has a bad performance in the case of the steering vector error and the polarization mismatch. What＇s more,it may even cause a signal cancellation phenomenon. In this paper,interference and noise covariance matrix reconstruction method is promoted into the quaternion field. We use Q-Capon polarization-angle spectrum estimation to reconstruct the interference and noise covariance matrix because it can help us get the interference and noise power. The method proposed not only can effectively avoid the diagonal loading method for solving the diagonal loading factor,but also can overcome the bad performance caused by the pointing error and the polarization mismatch. The computer simulation shows that the method has a better performance compared with other methods in the quaternion domain.

关 键 词：鲁棒自适应波束形成 电磁矢量传感器阵列 四元数 干扰加噪声协方差矩阵重构 

分 类 号：TN971.1[电子电信—信号与信息处理]