Title :
Stabilisation of fast QRD inverse-updates adaptive filtering algorithm
Author :
Sakai, H. ; Hori, S.
Author_Institution :
Dept. of Syst. Sci., Kyoto Univ., Japan
fDate :
4/1/2001 12:00:00 AM
Abstract :
A stabilisation method for the fast QRD inverse-updates adaptive filtering algorithm of Proudler (1994) involving modification by leakage is proposed. A stability analysis of the error propagation of the algorithm is presented using the averaging principle. It is shown that the proposed algorithm is stable if the variance of the one-step-ahead linear predictor of the input signal is less than that of the prediction error, with the forgetting factor and the leakage factor sufficiently close to 1, under additional assumptions on the leakage factor and the linear prediction coefficient
Keywords :
adaptive filters; adaptive signal processing; filtering theory; least squares approximations; matrix decomposition; matrix inversion; numerical stability; prediction theory; RLS adaptive filtering; averaging principle; error propagation; fast QRD inverse-updates adaptive filtering algorithm; forgetting factor; input signal; leakage factor; linear prediction coefficient; one-step-ahead linear predictor; prediction error; stabilisation method; stability analysis; stable algorithm;
Journal_Title :
Vision, Image and Signal Processing, IEE Proceedings -
DOI :
10.1049/ip-vis:20010153
