Multichannel blind deconvolution of non-minimum phase systems using information backpropagation

摘要: We present a novel method-filter decomposition approach, for multichannel blind deconvolution of non-minimum phase systems. In earlier work we developed an efficient natural gradient algorithm causal FIR filters. this paper further study the method noncausal decompose doubly finite filters into product two filters, filter and filter. The is employed to train filter, information backpropagation training Simulations are given illustrate effectiveness validity algorithm.