Semi-Blind Source Separation for Nonlinear Acoustic Echo Cancellation

System approach to robust acoustic echo cancellation through semi-blind source separation based on independent component analysis

Semi-Blind Source Parameter Separation

Independent Component Analysis (ICA) is a useful extension of standard Principal Component Analysis (PCA). The ICA model is utilized mainly in blind separation of unknown source signals from their linear mixtures. In some applications, the mixture coeecients are totally unknown, while some knowledge about temporal model exists. In this paper, we propose a learning system for semi-blind binary s...

Adaptive Noise Cancellation and Blind Source Separation

The use of adaptive noise cancellers (ANCs) to reduce the noise level prior to source separation is investigated in this paper. The foetal electrocardiogram (ECG) extraction problem in particular is addressed, which as well as with noise, is compounded by the non-stationary nature of the measurements. Consequently, computer simulations show that the combined Kalman filter and natural gradient a...

Blind and semi-blind source separation

Adaptive Volterra filters for nonlinear acoustic echo cancellation

More than 20 years ago it was shown that adaptive Volterra filters can be used for telephone echo cancellation, but the inherent need for a large number of coefficients has inhibited their practical use. Recently, first attempts have been made to combat nonlinearities in the acoustic echo path of hands-free telephones, which are caused by low-cost audio components. As these attempts are based o...