Conference paper
Adaptive Filtering for Non-Gaussian Processes
A new stochastic gradient robust filtering method, based on a non-linear amplitude transformation, is proposed. The method requires no a priori knowledge of the characteristics of the input signals and it is insensitive to the signals distribution and to the stationarity of the signals. A simulation study, applying both synthetic and real-world signals, shows that the proposed method has overall better robustness performance, in terms of modeling error, compared with state-of-the-art robust filtering methods.
A remarkable property of the proposed method is that it can handle double-talk in the acoustical echo-cancellation problem.
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2000 |
Pages: | 424-427 |
Proceedings: | Proceedings of International Conference on Acoustics, Speech and Signal Processing, ICASSP2000 |
ISBN: | 0780362934 and 9780780362932 |
ISSN: | 2379190x and 15206149 |
Types: | Conference paper |
DOI: | 10.1109/ICASSP.2000.861999 |
Adaptive algorithm Adaptive filters Digital filters Digital signal processing Echo cancellers FIR filters Filtering Finite impulse response filter Mathematical model Robustness Stochastic processes acoustic signal processing acoustical echo-cancellation adaptive FIR filter adaptive algorithms adaptive filtering adaptive filters adaptive signal processing audio signals double-talk echo suppression fractional lower order moments gradient methods modeling error nonGaussian processes nonlinear amplitude transformation real-world signals signal distribution simulation stochastic gradient robust filtering stochastic processes synthetic signals