Analysis of the QR-RLS algorithm for colored-input signals

Author

Diniz, Paulo S R ; Siqueira, Marcio G.

Author_Institution

COPPE/EE, Univ. Federal do Rio de Janeiro, Brazil

Volume

2

fYear

1995

fDate

9-12 May 1995

Firstpage

1428

Abstract

The recursive least squares (RLS) algorithms for FIR adaptive filtering are particularly attractive due to their fast convergence, especially for correlated input signals. A detailed analysis of the QR-RLS algorithm in finite and infinite precision implementations is presented, emphasizing the case where the input signal samples are correlated. The expressions for the mean square values of all internal variables in a steady state are first derived. These expressions are the key used to determine the dynamic range of the internal signals, and to derive the analytical expressions for the mean square values of the deviations in the output variables of the algorithm in finite wordlength implementations. Previous works address this problem considering the input signal a white noise, a situation not so often encountered in practice. The accuracy of all analytical results are verified through a number of computer simulations

Keywords

FIR filters; adaptive filters; adaptive signal processing; correlation methods; filtering theory; least squares approximations; recursive filters; signal sampling; FIR adaptive filtering; QR-RLS algorithm; colored-input signals; computer simulations; correlated input signal samples; dynamic range; fast convergence; finite precision implementation; finite wordlength implementations; infinite precision implementation; internal signals; internal variables; mean square values; output variables; recursive least squares algorithms; steady state; Adaptive filters; Algorithm design and analysis; Convergence; Dynamic range; Filtering algorithms; Finite impulse response filter; Least squares methods; Resonance light scattering; Signal analysis; Steady-state;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on

Conference_Location

Detroit, MI

ISSN

1520-6149

Print_ISBN

0-7803-2431-5

Type

conf

DOI

10.1109/ICASSP.1995.480551

Filename

480551