Region-Based Wavelet-Packet Adaptive Algorithm for Sparse Response Identification

This paper proposes a new wavelet-packet-based algorithm for sparse response identification. The distinctive features of the new algorithm are a region-based strategy for adaptive weight activation and a new deactivation/activation schedule across the transform scales. The new algorithm shows improved performance when compared to existing wavelet-based algorithms with similar characteristics. The new strategies lead to a faster wavelet-packet transform construction and to improved robustness to design parameters when compared to previous solutions. Monte Carlo simulation results show good performances regarding convergence speed and robustness to design parameter choice.

[1]  Ali H. Sayed,et al.  Fundamentals Of Adaptive Filtering , 2003 .

[2]  Odair A. Noskoski,et al.  Wavelet-Packet-Based Adaptive Algorithm for Sparse Impulse Response Identification , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[3]  Milos Doroslovacki,et al.  Proportionate adaptive algorithms for network echo cancellation , 2006, IEEE Transactions on Signal Processing.

[4]  Neil J. Bershad,et al.  Fast coupled adaptation for sparse impulse responses using a partial haar transform , 2005, IEEE Transactions on Signal Processing.

[5]  Akihiko Sugiyama,et al.  A fast convergence algorithm for sparse-tap adaptive FIR filters identifying an unknown number of dispersive regions , 2002, IEEE Trans. Signal Process..

[6]  Iven M. Y. Mareels,et al.  LS detection guided NLMS estimation of sparse systems , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[7]  Shannon D. Blunt,et al.  Rapid identification of a sparse impulse response using an adaptive algorithm in the Haar domain , 2003, IEEE Trans. Signal Process..

[8]  S. Mallat.  A wavelet tour of signal processing , 1998 .