Fractional-order LMS filter Based on Chaotic Particle Swarm

In the traditional LMS adaptive algorithm, not only is the step factor a fixed value, but there is also an irreconcilable conflict between the convergence speed and the steady-state error. Although fractional-order LMS algorithms have been proposed to improve the convergence characteristics, the ste...

Full description

Saved in:
Bibliographic Details
Published in2021 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST) pp. 74 - 78
Main Authors Wan, Honglin, Sheng, Hu, Yang, Jingxin
Format Conference Proceeding
LanguageEnglish
Published IEEE 10.12.2021
Subjects
Chaotic Particle Swarm Algorithm
Classification algorithms
component
Convergence speed and steady-state error
Filtering algorithms
Fractional order LMS algorithm
Heuristic algorithms
Particle swarm optimization
Sociology
Standard Particle Swarm
Steady-state
Technological innovation
Online AccessGet full text
DOI10.1109/IAECST54258.2021.9695741

Cover

More Information
Summary:In the traditional LMS adaptive algorithm, not only is the step factor a fixed value, but there is also an irreconcilable conflict between the convergence speed and the steady-state error. Although fractional-order LMS algorithms have been proposed to improve the convergence characteristics, the step factor is still a fixed value, which is prone to divergence when not set properly. In this paper, the fractional-order LMS algorithm is improved and combined with the chaotic particle swarm algorithm. The fractional-order LMS algorithm is combined with a chaotic particle swarm algorithm, and the fixed step factor value in the fractional-order LMS algorithm is changed to a dynamically varying step factor. The results show that the convergence speed and steady-state error of the fractional-order LMS filter are effectively improved by adding the chaotic particle swarm.
DOI:10.1109/IAECST54258.2021.9695741