Title :
A set-membership nlms algorithm with adaptive error bound
Author :
Bhotto, Md Zulfiquar Ali ; Antoniou, Andreas
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Victoria, Victoria, BC
Abstract :
The error-bound estimation problem in set-membership (SM) adaptive filtering is considered. A solution to this problem is proposed by incorporating an adaptive error bound in the SM normalized least-mean-square (NLMS) algorithm. The proposed algorithm retains the data-selective property and speed of convergence of SM algorithms without a significant increase in the computational complexity, and like other algorithms of the SM family it is much more efficient than the conventional NLMS algorithm.
Keywords :
adaptive estimation; adaptive filters; computational complexity; least mean squares methods; set theory; adaptive error-bound estimation problem; computational complexity; convergence speed; data-selective property; set-membership adaptive filtering; set-membership normalized least-mean-square algorithm; Adaptive filters; Algorithm design and analysis; Computational complexity; Computer errors; Convergence; Electronic mail; Error correction; Estimation error; Filtering algorithms; Samarium; Adaptive filters; NLMS algorithm; set-membership adaptive filtering;
Conference_Titel :
Electrical and Computer Engineering, 2008. CCECE 2008. Canadian Conference on
Conference_Location :
Niagara Falls, ON
Print_ISBN :
978-1-4244-1642-4
Electronic_ISBN :
0840-7789
DOI :
10.1109/CCECE.2008.4564897
