Algorithm to eliminate permutation of frequency domain blind source separation based on influence factor

作者：Bo, Xiang Lei^*; He, Yi Gang; Yin, Bai Qiang; Fang, Ge Feng; Fan, Xiao Teng; Li, Zhong Qun

来源：Acta Electronica Sinica, 2014, 42(2): 360-365.

DOI：10.3969/j.issn.0372-2112.2014.02.023

• Convolutive mixture
• Frequency domain blind source separation
• Frequency domains
• Frequency-domain BSS
• Influence factor
• Instantaneous mixtures
• Permutation
• Permutation problems

Blind source separation (BSS) for convolutive mixtures can be solved efficiently in the frequency domain, however, the permutation problem must be solved in the frequency domain BSS. This paper studies the influence of distance between frequency bins and separation quality at each frequency on permutation algorithm based on the amplitude correlation between neighbor bins, and proposes an improved method for permutation. The influence factor is utilized to control the influence of separation quality and distance between frequency bins on permutation. Lager influence factor is set for frequency bins of short distance and good separation quality; otherwise a smaller factor is set. Functions for setting influence factor are discussed in detail in this paper. Experiments of BSS for instantaneous mixtures, convolutive mixtures and real-room recorded data are made at last. The results demonstrate the performance of the proposed algorithm is good.

• 出版日期2014
• 单位合肥工业大学