For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Variable Tap-Length NLMS Algorithm with Adaptive Parameter
Yufei HAN Mingjiang WANG Boya ZHAO
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2017/08/01
Online ISSN: 1745-1337
Type of Manuscript: LETTER
Category: Digital Signal Processing
adaptive filter, variable tap-length algorithm, Sigmoid limited function, convergence speed, robustness to noise,
Full Text: PDF(289.4KB)
>>Buy this Article
Improved fractional variable tap-length adaptive algorithm that contains Sigmoid limited fluctuation function and adaptive variable step-size of tap-length based on fragment-full error is presented. The proposed algorithm can solve many deficiencies in previous algorithm, comprising small convergence rate and weak anti-interference ability. The parameters are able to modify reasonably on the basis of different situations. The Sigmoid constrained function can decrease the fluctuant amplitude of the instantaneous errors effectively and improves the ability of anti-noise interference. Simulations demonstrate that the proposed algorithm equips better performance.