A permutation algorithm of frequency-domain blind sourceseparation based on influence weights

Abstract

In the frequency domain, convolutive mixes with blind source separation can be successfully resolved. But thepermutation issue in frequency-domain blind source separation needs to be resolved. We investigated the impactof frequency space and separation performance at each frequency bin on the amplitude correlation permutationalgorithm with the goal of addressing the permutation ambiguity problem in frequency-domain blind sourceseparation of convolutive mixtures, and we proposed an enhanced permutation algorithm. The improved algorithmuses spacing influence weight and performance influence weight to control the influence of the frequency binssorted in the neighborhood on the frequency bins unsorted. Experiments have shown that the two influenceweights are effective. Finally, blind source separation experiments are performed on the speech signals under thetwo convolutive mixing models and the simulated room mixing model. According to experiments, the increasedsignal to interference plus noise ratio of separated signals demonstrates that the improved algorithm outperformsthe amplitude correlation permutation algorithm in terms of separation performance and robustness.