Title :
RLS lattice algorithm using gradient based variable forgetting factor
Author :
So, C.F. ; Ng, S.C. ; Leung, S.H.
Author_Institution :
Dept. of Electr. Eng., Hong Kong Polytech. Univ., China
Abstract :
A gradient based variable forgetting factor (GVFF) RLS lattice (RLSL) algorithm is introduced in this paper. The steepest descent approach is used to control the forgetting factor which is based on the dynamic equation of the gradient of the mean square error. Compared with the standard RLSL algorithm, GVFF-RLSL algorithm gives fast tracking with a small mean square model error and its performance is not degraded much even in low signal-to-noise ratios (SNR) for time varying systems.
Keywords :
adaptive signal processing; gradient methods; lattice filters; least squares approximations; mean square error methods; recursive estimation; recursive filters; time-varying systems; dynamic equation; fast tracking; forgetting factor control; gradient based variable forgetting factor; low signal-to-noise ratios; recursive least squares lattice algorithm; steepest descent approach; time varying systems; Computational complexity; Degradation; Equations; Filters; Lattices; Least squares methods; Mean square error methods; Resonance light scattering; Steady-state; Time varying systems;
Conference_Titel :
Neural Networks, 2003. Proceedings of the International Joint Conference on
Print_ISBN :
0-7803-7898-9
DOI :
10.1109/IJCNN.2003.1223857
