Low-delay frequency domain LMS algorithm
- 1 January 1992
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4 (15206149) , 9-12 vol.4
- https://doi.org/10.1109/icassp.1992.226424
Abstract
The authors deal with frequency-domain (FD) adaptive identification with application to acoustic echo cancellation. Several proposed frequency-domain LMS algorithms are reviewed and unified in the common framework of fast convolution using the short-term Fourier transform. A more flexible identification scheme is then proposed, making it possible to independently control the delay and the adaptation rate. The theoretical proof of convergence of this new scheme is provided together with simulations demonstrating the efficiency of this method in both stationary and nonstationary environments.Keywords
This publication has 6 references indexed in Scilit:
- Multidelay block frequency domain adaptive filterIEEE Transactions on Acoustics, Speech, and Signal Processing, 1990
- Learning characteristics of stochastic-gradient-descent algorithms: A general study, analysis, and critiqueSignal Processing, 1984
- Unconstrained frequency-domain adaptive filterIEEE Transactions on Acoustics, Speech, and Signal Processing, 1982
- Fast implementations of LMS adaptive filtersIEEE Transactions on Acoustics, Speech, and Signal Processing, 1980
- Time-frequency representation of digital signals and systems based on short-time Fourier analysisIEEE Transactions on Acoustics, Speech, and Signal Processing, 1980
- On the asymptotic eigenvalue distribution of Toeplitz matricesIEEE Transactions on Information Theory, 1972