Source separation based on coupled single DOA estimation processors

The separation of sources from an array of sensors is addressed. Based on Kolmogorov´s theorem, an adaptive algorithm is proposed to separate simultaneously all the unknown impinging sources on an aperture of sensors. This method solves the problem of simultaneous direction of arrival (angle of arrival) estimation and source beamforming in narrowband array processing. Kolmogorov´s theorem is implemented by a network composed of two different stages: the inhibition and the optimization stages.<>