New adaptive algorithms based on multi-band decomposition of the error signal

New adaptive algorithms based on multi-band decomposition of the error signal and application of a different convergence factor for each band are derived. With this approach, tracking ability and performance in steady state can be traded off along the frequency domain giving rise to estimates of the adaptive filter coefficients closer to the ideal response as compared to those obtained with conventional least-mean-square (LMS) and recursive least-squares (RLS) algorithms, particularly when the statistical properties of the analyzed signal vary along the frequency spectrum. A new adaptation technique for the forgetting factor depending exclusively on the autocorrelation values of the input signal is also introduced and a multi-band RLS algorithm, with an independent variable forgetting factor for each band, suitable for the analysis of nonstationary signals is described. Computer experiments comparing the performance of multi-band and conventional LMS and RLS algorithms are shown.

[1]  Mineo Kaneko,et al.  Adaptive AR spectral estimation based on multi-band decomposition of the linear prediction error with variable forgetting factors , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Branko D. Kovacevic,et al.  Robust recursive AR speech analysis , 1995, Signal Process..

[3]  K. Tokuda,et al.  Multi - Band Decomposition of the Linear Prediction Error Applied to Adaptive AR Spectral Estimation , 1997 .

[4]  Keiichi Tokuda,et al.  AR spectrum estimation based on wavelet representation , 1994, Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94.

[5]  Paulo S. R. Diniz,et al.  Adaptive Filtering: Algorithms and Practical Implementation , 1997 .

[6]  J. Brossier.  Egalisation adaptative et estimation de phase : application aux communications sous-marines , 1992 .

[7]  Thomas Kailath,et al.  Fast transversal filters with data sequence weighting , 1989, IEEE Trans. Acoust. Speech Signal Process..

[8]  K. Tokuda,et al.  Adaptive AR Spectral Estimation Based on Wavelet Decomposition of the Linear Prediction Error , 1996 .

[9]  Simon Haykin,et al.  Adaptive filter theory (2nd ed.) , 1991 .

[10]  Pierre Priouret,et al.  Adaptive Algorithms and Stochastic Approximations , 1990, Applications of Mathematics.

[11]  H. Kushner,et al.  Analysis of adaptive step-size SA algorithms for parameter tracking , 1995, IEEE Trans. Autom. Control..

[12]  Edward J. Powers,et al.  Time-varying spectral estimation using AR models with variable forgetting factors , 1991, IEEE Trans. Signal Process..

[13]  K. Tokuda,et al.  Multi-band decomposition of the linear prediction error applied to the least-mean-square method with fixed and variable step-sizes , 1997, Proceedings of 1997 IEEE International Symposium on Circuits and Systems. Circuits and Systems in the Information Age ISCAS '97.

[14]  Keiichi Tokuda,et al.  RLS algorithms for adaptive AR spectrum analysis based on multi-band decomposition of the linear prediction error , 1996, Proceedings of Digital Processing Applications (TENCON '96).

[15]  M. Kaneko,et al.  AR spectral estimation based on multi-window analysis of the linear prediction error , 1995, 38th Midwest Symposium on Circuits and Systems. Proceedings.

[16]  Truong Q. Nguyen,et al.  Linear phase paraunitary filter banks: theory, factorizations and designs , 1993, IEEE Trans. Signal Process..

[17]  Peter Strobach.  Recursive triangular array ladder algorithms , 1991, IEEE Trans. Signal Process..

[18]  Subbarayan Pasupathy,et al.  Tracking improvements in fast RLS algorithms using a variable forgetting factor , 1988, IEEE Trans. Acoust. Speech Signal Process..