Title :
Fast adaptive algorithms for multichannel LS filtering
Author :
Glentis, G.O. ; Kalouptsidis, N.
Author_Institution :
Dept. of Phys., Athens Univ., Greece
Abstract :
Fast recursive least-squares (FRLS) algorithms for multichannel finite-impulse response (FIR) filtering and system identification are discussed. The exact case is considered where the predictor depends linearly on the parameters. Using suitable order updating procedures several distinct step-up, step-down or step-down, step-up schemes for the computation of the Kalman gain are presented. Single block-order as well as multiple single-order one-term recursions are developed
Keywords :
Kalman filters; digital filters; filtering and prediction theory; least squares approximations; Kalman gain; fast recursive least squares single block-order; multichannel LS filtering; multichannel finite-impulse response; multiple single-order one-term recursions; order updating procedures; step-down; step-up; system identification; Adaptive algorithm; Adaptive filters; Filtering; Finite impulse response filter; Kalman filters; Linear systems; Nonlinear filters; Partitioning algorithms; Physics; Vectors;
Conference_Titel :
Circuits and Systems, 1990., IEEE International Symposium on
Conference_Location :
New Orleans, LA
DOI :
10.1109/ISCAS.1990.112200
