Overdetermined Blind Source Separation: Using More Sensors Than Source Signals In A Noisy Mixture
M. Joho, H Mathis, Russell H. Lambert · 2000
This paper addresses the blind source separation problem for the case where more sensors than source signals are available. A noisy-sensor model is assumed. The proposed algorithm comprises two stages, where the first stage consists of a principal component analysis (PCA) and the second one of an independent component analysis (ICA). The purpose of the PCA stage is to increase the input SNR of the succeeding ICA stage and to reduce the sensor dimensionality. The ICA stage is used to separate the remaining mixture into its independent components. A simulation example demonstrates the performance of the algorithm proposed. 1. INTRODUCTION 1.1. Problem description Blind source separation (BSS) is a problem posed by many applications related to acoustics or communications. Usually the BSS problem is analyzed for the case where there are just as many sensors as source signals. Furthermore, ideal sensors are usually assumed, which have no additive sensor noise. Only little work has been don...