Title :
A robust statistics based adaptive lattice-ladder filter in impulsive noise
Author :
Zou, Yue-Xian ; Chan, Shing-Chow ; Ng, Tung-Sang
Author_Institution :
Dept. of Electr. & Electron. Eng., Hong Kong Univ., China
Abstract :
In this paper, a new robust adaptive lattice-ladder filter for impulsive noise suppression is proposed. The filter is obtained by applying the non-linear filtering technique reported by Kim and Efron (1995) and the robust statistic approach to the gradient adaptive lattice filter. A systematic method is also developed to determine the corresponding threshold parameters for impulse suppression. Simulation results showed that the performance of the proposed algorithm is better than the conventional RLS, N-RLS, the gradient adaptive lattice normalised-LMS (GAL-NLMS), RMN and ATNA algorithms when the input and desired signals are corrupted by individual and consecutive impulses. The initial convergence, steady-state error, computational complexity and tracking capability of the proposed algorithm are also comparable to the conventional GAL-NLMS algorithm
Keywords :
adaptive filters; convergence; filtering theory; impulse noise; interference suppression; ladder filters; lattice filters; nonlinear filters; parameter estimation; statistics; adaptive lattice-ladder filter; computational complexity; gradient adaptive lattice filter; impulsive noise suppression; initial convergence; nonlinear filtering technique; robust statistics based adaptive filter; steady-state error; threshold parameters; tracking capability; Adaptive filters; Computational complexity; Computational modeling; Convergence; Filtering; Lattices; Noise robustness; Resonance light scattering; Statistics; Steady-state;
Conference_Titel :
Circuits and Systems, 2000. Proceedings. ISCAS 2000 Geneva. The 2000 IEEE International Symposium on
Conference_Location :
Geneva
Print_ISBN :
0-7803-5482-6
DOI :
10.1109/ISCAS.2000.856116
