Title :
Tracking capabilities of the LMS adaptive filter in the presence of gain variations
Author :
Bergmans, Jan W M
Author_Institution :
Hitachi Ltd., Tokyo, Japan
fDate :
4/1/1990 12:00:00 AM
Abstract :
The tracking capabilities of the least mean square (LMS) adaptive filter when it is used to identify a nonstationary channel suffering from gain variations are studied. An analytical solution for the steady-state lag error is obtained, and its validity is verified by means of simulations. Depending on the magnitude and bandwidth of the variations, the total steady-state error is found either to increase monotically as a function of the adaptation parameter μ or to exhibit a global minimum for a finite, nonzero value of μ
Keywords :
adaptive filters; least squares approximations; tracking; LMS adaptive filter; adaptation parameter; bandwidth; gain variations; global minimum; least mean square; magnitude; nonstationary channel; simulations; steady-state lag error; Adaptive filters; Additive noise; Analytical models; Bandwidth; Error analysis; Fluctuations; Least squares approximation; Steady-state; White noise; Working environment noise;
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
