一种解决盲分离和盲信号去卷积的信息最大化方法-毕业外文翻译.docVIP

英文原文 An information-maximisation approach to blind separation and blind deconvolution Anthony J. Bell and Terrence J. Sejnowski Computational Neurobiology Laboratory The Salk Institute 100010 N.Torrey Pines Road La Jolla, California 92037 Abstract We derive a new self-organising learning algorithm which maximises the infor- mation transferred in a network of non-linear units. The algorithm does not assume any knowledge of the input distributions, and is defined here for the zero-noise limit. Under these conditions, information maximisation has extra properties not found in the linear case(Linsker 1989). The non-linearities in the transfer function are able to pick up higher-order moments of the input distributions and perform something akin to true redundancy reduction between units in the output representation. This enable- s the network to separate statistically independent components in the inputs: a higher -order generalisation of Principal Components Analysis. We apply the network to the source separation (or cocktail party)problem, succe- ssfully separating unknown mixtures of up to ten speakers. We also show that a vari- ant on the network architecture is able to perform blind deconvolution (cancellation of unknown echoes and reverberation in a speech signal). Finally, we derive depend- encies of information transfer on time delays. We suggest that information maximi- sation provides a unifying framework for problems in ‘blind’ signal processing. 1 Introduction This paper presents a convergence of two lines of research. The first, the developme- nt of information theoretic unsupervised learning rules for neural networks has been pioneered by Linsker 1992, Becker Hinton 1992, Atick Redlich 1993, Plumble- y Fallside 1988 and others. The second is the use,in signal processing, of higher-order statistics, for separating out mixtures of independent sources (blind sep- aration) or reversing the effect of an unknown filter (blind deconvolution). Methods exist fo