Self-adaptive neural networks for blind separation of sources

Novel on-line learning algorithms with self adaptive learning rates (parameters) for blind separation of signals are proposed. The main motivation for development of new learning rules is to improve convergence speed and to reduce cross-talk, especially for non-stationary signals. Furthermore, we have discovered that under some conditions the proposed neural network models with associated learning algorithms exhibit a random switch of attention, i.e. they have the ability of chaotic or random switching or cross-over of output signals in such way that a specified separated signal may appear at various outputs at different time windows. Validity, performance and dynamic properties of the proposed learning algorithms are investigated by computer simulation experiments.