Title :
A simple model for the effect of normalization on the convergence rate of adaptive filters
Author :
Nascimento, Vítor H.
Author_Institution :
Electron. Syst. Eng. Dept., Sao Paulo Univ., Brazil
Abstract :
We propose a new simple model for the input regressor vectors in adaptive filters. This model allows more insight on the effect of normalization on the convergence rate and eigenvalue spread for the normalized least-mean-squares algorithm (NLMS). Using the new model, we show that NLMS works best to reduce eigenvalue spread when the input regressor vector points to all directions with equal probability, but with direction-dependent power.
Keywords :
adaptive filters; convergence of numerical methods; eigenvalues and eigenfunctions; filtering theory; least mean squares methods; probability; NLMS; adaptive filter convergence rate; direction-dependent power; eigenvalue spread; input regressor vectors; normalization; normalized LMS algorithm; normalized least-mean-squares algorithm; regressor vector; Adaptive filters; Algorithm design and analysis; Autocorrelation; Brazil Council; Convergence; Eigenvalues and eigenfunctions; Equations; Least squares approximation; Stability;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on
Print_ISBN :
0-7803-8484-9
DOI :
10.1109/ICASSP.2004.1326292
