基于高阶累积量的空间特征估计方法及其应用

以均匀直线阵为基础,提出了一种基于阵列输出四阶累积量的信号空间特征估计方法,并将其用于相干多径环境下的多用户信号空间特征的估计。利用阵列输出四阶累积量构造了一种空间特征矩阵,证明了对其作特征分解可以得到各用户的空间特征。然后在空间特征估计基础上,结合前向/共轭后向空间平滑技术对相干多径的波达方向(DoA)进行了估计。该算法是一种不依赖于信号具体特征的全盲估计方法,适用于任意加性高斯噪声,且各DoA与信源自动配对。理论分析和仿真结果说明了算法的有效性和稳健性。

Spatial signature estimation algorithm based on high order cumulant and its application

Based on uniform linear array and fourth order cumulant of array output signal,a novel algorithm for spatial signature estimation is presented and applied to estimate multi-user spatial signature in the coherent multi-path environment.Using the fourth order cumulant of array output,a spatial signature matrix is constructed,which is proved to be effective to estimate multi-user spatial signature via eigen-decomposition.Then,based on spatial signature estimation and forward/conjugate backward spatial smoothing technique,the direction-of-arrival(DoA) of coherent multple paths is estimated.The algorithm,which is a fully-blind,estimation method independent of the specific characteristic of the signals,is suitable to arbitrary additive Gaussian noise,and each DoA can automatically match the sources.Theoretical analysis and computer simulation prove the effectiveness and robustness of the algorithm.