ABSTRACT
Independent component analysis (ICA) is a signal processing and data analysis technology that was first introduced in the late 1980s. ICA is a multivariate analytical approach that aims at the retrieval of an ensemble of independent source signals out of an ensemble of linear mixtures. It can achieve signal separation by determining the inverse of the mixing process blindly, without prior knowledge of the original source signals or the mixing process. In the past few years, blind source separation by ICA has received significant interest in biomedical signal processing and interpretation such as the analysis of electroencephalogram (EEG, Makeig, Bell, Jung & Sejnowski, 1996) and functional magnetic resonance imaging (fMRI, McKeown, Makeig, Brown, Jung, Kinderman & Sejnowski, 1998) data.
