Open AccessAffine projection Weibull M‐transform least mean square algorithm against impulsive interference
Author(s) -
Liu Pingchuan,
Fan Kuangang,
Qiu Haiyun
Publication year - 2021
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/cmu2.12249
Subject(s) - algorithm , affine transformation , least mean squares filter , impulse noise , affine combination , computer science , weibull distribution , norm (philosophy) , mathematics , adaptive filter , artificial intelligence , statistics , pixel , pure mathematics , law , political science
The well‐known affine projection sign algorithm is one of the classic adaptive filtering methods for denoising and channel equalisation, which can achieve robust performance against coloured input and impulse noise. This work proposes the affine projection Weibull M‐transform least mean square algorithm to improve the steady‐state performance of affine projection algorithms. The proposed algorithm combines the Weibull M‐transform cost function with the L2‐norm constraint of weight vector to improve the performance of existing affine projection algorithms and achieve a better convergence rate and better misalignment. This novel algorithm also preserves good performance against impulsive interference and coloured input. Simulation results demonstrated that the proposed algorithm has a remarkable improvement in convergence performance compared with other algorithms.