Abstract
Independent Component Analysis (ICA) is one of the emerging areas in the field of neural networks and signal processing. It is used to extract independent signal components from their observed linear mixtures at an array of sensors. This technique is also known as Blind Source Separation (BSS). Various statistical techniques based on information theoretic and algebraic approaches exist for performing ICA. Neural algorithms derived from these approaches are used for separating the independent components. In this paper we have used an objective function based on the independence criterion of the signals. Minimization of this objective function using Edgeworth expansion of probability density function yields a nonlinear function for the source separation algorithm. Performance of the signals for artificially generated signals has been evaluated.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have