Title :
A non-stationary RLS algorithm for adaptive tracking of Markov time varying channel
Author :
Alouane, M. Turki-Hadaj ; Jaidane-Saidane, M.
Author_Institution :
ENIT, L.S. Telecoms, Tunis, Tunisia
Abstract :
We propose a new adaptive algorithm designed to track system presented by a filter that has Markovian time evolution. As the non-stationary LMS (NSLMS) algorithm, the non-stationary RLS (NSRLS) algorithm performs better than the LMS and is able to identify the unknown order and parameters of the Markov model. However in the case of the NSRLS algorithm, the convergence speed of the Markovian parameter is very high compared to that of the NSLMS algorithm. Moreover, the NSRLS algorithm has a better tracking capacity than the NSLMS, especially when the filter poles that characterize time variations of the channel are close to the unit circle
Keywords :
Markov processes; adaptive filters; adaptive signal processing; filtering theory; least squares approximations; poles and zeros; recursive filters; time-varying channels; tracking filters; Markov model; Markov time varying channel; Markovian parameter; Markovian time evolution; NSLMS algorithm; NSRLS algorithm; adaptive algorithm; adaptive filter; adaptive tracking; convergence speed; filter poles; nonstationary RLS algorithm; unit circle; Adaptive algorithm; Adaptive filters; Algorithm design and analysis; Convergence; Kalman filters; Least squares approximation; Resonance light scattering; Statistics; Steady-state; Time varying systems;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1999. Proceedings., 1999 IEEE International Conference on
Conference_Location :
Phoenix, AZ
Print_ISBN :
0-7803-5041-3
DOI :
10.1109/ICASSP.1999.756210
