A Unifying View on Blind Source Separation of Convolutive Mixtures Based on Independent Component Analysis

In many daily-life scenarios, acoustic sources recorded in an enclosure can only be observed with other interfering sources. Hence, convolutive Blind Source Separation (BSS) is a central problem in audio signal processing. Methods based on Independent Component Analysis (ICA) are especially important in this field as they require only few and weak assumptions and allow for blindness regarding the original source signals and the acoustic propagation path. Most of the currently used algorithms belong to one of the following three families: Frequency Domain ICA (FD-ICA), Independent Vector Analysis (IVA), and TRIple-N Independent component analysis for CONvolutive mixtures (TRINICON). While the relation between ICA, FD-ICA and IVA becomes apparent due to their construction, the relation to TRINICON is not well established yet. This paper fills this gap by providing an in-depth treatment of the common building blocks of these algorithms and their differences, and thus provides a common framework for all considered algorithms.

Automated summary

In many scenarios, acoustic sources recorded in an enclosure are often observed together with interfering sources, making convolutive Blind Source Separation (BSS) a crucial problem in audio signal processing. Methods based on Independent Component Analysis (ICA) are particularly significant as they have minimal assumptions and allow for blindness in the original source signals and acoustic propagation path. This paper aims to establish a comprehensive framework by exploring the common building blocks and differences between various algorithms, including Frequency Domain ICA (FD-ICA), Independent Vector Analysis (IVA), and TRIple-N Independent component analysis for CONvolutive mixtures (TRINICON), thus bridging the gap in understanding the relation to TRINICON.