Modifications of FastICA in Convolutive Blind Source Separation
This work addresses convolutive blind source separation, a domain-specific problem in signal processing, but appears incremental as it modifies an existing method (FastICA) rather than introducing a new paradigm.
The paper tackles the difficulties of spatial-temporal prewhitening and para-unitary filters in convolutive blind source separation by proposing modifications to FastICA, resulting in a method that performs simple prewhitening and optimizes contrast functions under diagonalization constraints, with performance verified through numerical simulations.
Convolutive blind source separation (BSS) is intended to recover the unknown components from their convolutive mixtures. Contrary to the contrast functions used in instantaneous cases, the spatial-temporal prewhitening stage and the para-unitary filters constraint are difficult to implement in a convolutive context. In this paper, we propose several modifications of FastICA to alleviate these difficulties. Our method performs the simple prewhitening step on convolutive mixtures prior to the separation and optimizes the contrast function under the diagonalization constraint implemented by single value decomposition (SVD). Numerical simulations are implemented to verify the performance of the proposed method.