基于非参数密度估计的多通道盲解卷积

非参数密度估计方法被用来直接估计在自然梯度盲解郑积算法中遇到的评价函数(score function)。与用一个非线性函数简单地代替评价函数相比较,这种直接估计评价函数的方法的主要优点是:它可以用来对杂系混合信号,即同时包含超高斯和亚高斯的信号,进行盲解卷积。因为评价函数可以被直接的估计出来,因此,就不需要针对不同的源信号选择不同的非线性函数来代替评价函数。这种方法可以用在更加“盲”的情况。

Multichannel Blind Deconvolution Based on Nonparametric Density Estimation

Nonparametric density estimation method is used to estimate directly the score functions encountered in relative gradient adaptation algorithms in multichannel blind deconvolution problem. Comparing with the method of simple nonlinear functions replacement, the key advantage of the direct estimation of the score functions lies in the tact that it enables the algorithms to separate hybrid mixtures of sources that contain both super-Gaussian and sub-Gaussian signals successfully. The source statistics required for the choices of the nonlinear functions is no longer needed since the score functions are estimated directly. The algorithms is thus expected to be applicable to more "blind" cases. The performances are evaluated in this paper to show the effectiveness of this method.