ADAPTIVE BLIND SIGNAL-PROCESSING - NEURAL-NETWORK APPROACHES

Learning algorithms and underlying basic mathematical ideas are presen ted for the problem of adaptive blind signal processing, especially in stantaneous blind separation and multichannel blind deconvolution/equa lization of independent source signals. We discuss recent developments of adaptive learning algorithms based on the natural gradient approac h and their properties concerning convergence, stability, and efficien cy. Several promising schemas ale proposed and reviewed in the paper. Emphasis is given to neural networks or adaptive filtering models and associated online adaptive nonlinear learning algorithms. Computer sim ulations illustrate the performance of the developed algorithms. Some results presented in this paper are new and are being published for th e first time.