Title :
Adaptive Volterra filtering using M-band wavelet transform and Gram-Schmidt orthogonalisation
Author :
Kang, D.J. ; Hwang, D.O. ; Nam, S.W. ; Powers, E.J.
Author_Institution :
Div. of Electr. & Comput. Eng., Hanyang Univ., Seoul, South Korea
fDate :
3/14/2002 12:00:00 AM
Abstract :
A new adaptive Volterra filter with fast convergence is proposed, where the M-band discrete wavelet transform and Gram-Schmidt orthogonalisation are utilised. In particular, the cascade combination of two such procedures reduces the eigenvalue spread of the Volterra input auto-correlation matrix, thus improving the convergence speed of the adaptive nonlinear filtering
Keywords :
Volterra series; adaptive filters; convergence of numerical methods; correlation theory; discrete wavelet transforms; eigenvalues and eigenfunctions; filtering theory; matrix algebra; nonlinear filters; Gram-Schmidt orthogonalisation; M-band discrete wavelet transform; adaptive Volterra filter; cascade combination; convergence speed; eigenvalue spread; input auto-correlation matrix; nonlinear filtering;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20020190
