Title :
Stationary points of the recursive generalized least squares algorithm for adaptive notch filtering
Author :
DragoSeviC, Manna V.
Author_Institution :
Comput. Syst. Design Lab., Inst. of Nucl. Sci. ´´Vinca´´, Belgrade, Yugoslavia
fDate :
4/1/1993 12:00:00 AM
Abstract :
Possible convergence points of the generalized least squares adaptive notch filtering algorithm are analytically derived for the multiple sinusoid case, showing explicitly the dependence of the asymptotic bias on the pole contraction factor, signal-to-noise ratio, and the true model parameters. The results for symmetric and nonsymmetric parameterization are compared
Keywords :
adaptive filters; convergence of numerical methods; filtering and prediction theory; least squares approximations; notch filters; adaptive notch filtering; asymptotic bias; convergence points; multiple sinusoid case; nonsymmetric parameterization; pole contraction factor; recursive generalized least squares algorithm; signal-to-noise ratio; stationary points; symmetric parameterization; Adaptive filters; Adaptive signal processing; Filtering algorithms; Least squares approximation; Least squares methods; Recursive estimation; Signal processing; Signal processing algorithms; Source separation; Speech processing;
Journal_Title :
Signal Processing, IEEE Transactions on
