A Blind Signal Separation Technique using Combination of Second-Order and Higher-Order Approaches

In this paper, a novel technique to blind signal separation based on combination of the second order and the higher order approach is introduced. We transfer the problem of blind signal separation to the wavelet domain. The main idea behind this approach is that the mixing signal can be decomposed into sum of uncorrelated and/or independent subbands using the wavelet transform. In the beginning, the observed signal is pre-whitened in the time domain then, the initial separation matrix will be estimated from second order statistics decorrelation method in the wavelet domain. The estimating matrix will be used as initial separating matrix in the higher order statistics method to estimate the final separation matrix. The algorithm is tested using different examples different examples (chirp signals, sound signals and natural images). Our extensive experiments have confirmed that the use of the proposed procedure provides promising results in separating the image from noisy mixtures of image.