Independent Component Analysis Techniques for Blind Source Separation

Abstract

Blind Source Separation (BSS) is a technique for separating signals from unknown sources received at different sensors. It is assumed that every sensor gets a linear mixture of the unknown independent sources. One of the most important approaches for achieving source separation is Independent Component Analysis (ICA). newlineResearch work in BSS gained impetus in the early nineties. Recently, it has found many applications in the areas of speech processing, biomedical signal processing, feature extraction, telecommunications, digital image processing, financial time series analysis etc. Due to its wide-ranging applications, research interest in this field is continuously growing. newlineIn this research work, various ICA techniques have been proposed. Performance of these techniques has been evaluated with various artificially generated and real world signals. Comparison with other techniques indicates effectiveness of the proposed methods. newline