Title :
Transform domain approximate QR-LS adaptive filtering algorithm
Author :
Xin-Xing, Yang ; Chan, S.C.
Author_Institution :
Dept. of Electr. & Electron. Eng., Hong Kong Univ., China
Abstract :
An improved approximate QR-least squares (A-QR-LS) algorithm, called the transform domain A-QR-LS (TA-QR-LS) algorithm, is introduced. The input signal vector is approximately decorrelated by some unitary transformations before applying the A-QR-LS, which is shown to improve considerably the convergence speed of the A-QR-LS algorithm recently proposed by Liu (IEEE Trans. SP, vol. 43, pp. 720-729, 1995). Furthermore, it is possible to reduce the arithmetic complexities (O(N)) of the A-QR-LS, and TA-QR-LS algorithms by using Givens rotations instead of the Householder transformation. Simulation results show that the proposed TA-QR-LS algorithm is a good alternative to the conventional recursive least squares (RLS) algorithm in adaptive filtering applications involving multiple channels, acoustic modeling, and fast parameter variations.
Keywords :
adaptive filters; computational complexity; convergence; filtering theory; least squares approximations; recursive estimation; transforms; Givens rotations; Householder transformation; RLS algorithm; TA-QR-LS algorithms; acoustic modeling; adaptive filtering applications; approximate QR-least squares algorithm; arithmetic complexities; convergence speed; fast parameter variations; input signal vector; multiple channels; recursive least squares algorithm; simulation; transform domain A-QR-LS algorithm; transform domain approximate QR-LS adaptive filtering algorithm; unitary transformations; Adaptive filters; Arithmetic; Convergence; Decorrelation; Filtering algorithms; Least squares approximation; Least squares methods; Matrix decomposition; Parameter estimation; Vectors;
Conference_Titel :
Circuits and Systems, 2003. ISCAS '03. Proceedings of the 2003 International Symposium on
Print_ISBN :
0-7803-7761-3
DOI :
10.1109/ISCAS.2003.1205851
