基于二阶统计量的两步自适应盲分离算法

Adaptive Blind Source Separation Algorithm by Second-Order Statistics

提出了一种新的两步自适应盲分离算法,利用代价函数的极值点特性分别获得混合信号和白化信号的特征向量矩阵,实现自适应盲分离过程.该算法避免了离线计算所采用的特征值分解等复杂运算,并且只利用信号的二阶统计量信息,计算工作量低.仿真结果表明,对含有附加噪声的盲混合信号具有良好的分离效果.

A new two-steps adaptive blind source separation algorithm was presented. The eigenvector matrices of the mixed signals and the whitened signals were obtained by using the stationary point property of a special cost function. The algorithm avoids the singular decomposition that commonly used in off-line calculation, and only the second order statistics of the signals is used, so the algorithm has low computing cost. The simulations show that the algorithm can get good separation performance to the blind mixed signals with additive noises.