Blind separation of convolutive mixtures of cyclostationary sources using an extended natural gradient method

An on-line adaptive blind source separation algorithm for the separation of convolutive mixtures of cyclostationary source signals is proposed. The algorithm is derived by applying natural gradient iterative learning to the novel cost function which is defined according to the wide sense cyclostationarity of signals. The efficiency of the algorithm is supported by simulations, which show that the proposed algorithm has improved performance for the separation of convolved cyclostationary signals in terms of convergence speed and waveform similarity measurement, as compared to the conventional natural gradient algorithm for convolutive mixtures.