Title :
Comparison of based adaptive predictive schemes for improvement of tracking randomly time-varying systems
Author :
Jebaral, S.B. ; Jaidane-Saidane, M.
Author_Institution :
LS Telecoms, Campus Univ., Le Belvedere, Tunisia
Abstract :
This paper presents a comparison between three based adaptive predictive schemes used in order to improve the tracking capability of the LMS algorithm. We identify system variations modeled by a random walk. Using a theoretical analysis and simulation results, we illustrate the contribution of coupled adaptive prediction and system identification for highly correlated stationary inputs and nonstationary (speech) inputs
Keywords :
identification; least mean squares methods; prediction theory; random processes; time-varying systems; tracking; LMS algorithm; based adaptive predictive scheme; highly correlated stationary inputs; nonstationary inputs; randomly time-varying systems; speech inputs; system identification; tracking capability; Adaptive filters; Additive noise; Analytical models; Convergence; Filtering algorithms; Least squares approximation; Predictive models; Speech analysis; System identification; Time varying systems;
Conference_Titel :
Electronics, Circuits and Systems, 1998 IEEE International Conference on
Conference_Location :
Lisboa
Print_ISBN :
0-7803-5008-1
DOI :
10.1109/ICECS.1998.813361
