ScholarLens Back to search
2009•The Journal of the Acoustical Society of KoreaRequires access

Variable Step Size LMS Algorithm Using the Error Difference

Hong-Chae Woo

Open publisher page 1 citations

Abstract

In communications and signal processing area, a number of least mean square adaptive algorithms have been used because of simplicity and robustness. However the LMS algorithm is known to have slow and non-uniform convergence. Various variable step size LMS adaptive algorithms have been introduced and researched to speed up the convergence rate. A variable step size LMS algorithm using the error difference for updating the step size is proposed. Compared with other algorithms, simulation results show that the proposed LMS algorithm has a fast convergence. The theoretical performance of the proposed algorithm is also analyzed for the steady state.

About this research paper

What this paper is about

In communications and signal processing area, a number of least mean square adaptive algorithms have been used because of simplicity and robustness. However the LMS algorithm is known to have slow and non-uniform convergence. Various variable step size LMS adaptive algorithms have been introduced and researched to speed up the convergence rate. A variable step size LMS algorithm using the error difference for updating the step size is proposed. Compared with other algorithms, simulation results show that the proposed LMS algorithm has a fast convergence. The theoretical performance of the proposed algorithm is also analyzed for the steady state.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

In communications and signal processing area, a number of least mean square adaptive algorithms have been used because of simplicity and robustness. However the LMS algorithm is known to have slow and non-uniform convergence. Various variable step size LMS adaptive algorithms have been introduced and researched to speed up the convergence rate. A variable step size LMS algorithm using the error difference for updating the step size is proposed. Compared with other algorithms, simulation results show that the proposed LMS algorithm has a fast convergence. The theoretical performance of the proposed algorithm is also analyzed for the steady state.

Key concepts: Least mean squares filter, Robustness (evolution), Algorithm, Convergence (economics), Rate of convergence, Variable (mathematics), Mathematics, Adaptive filter

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Variable Step Size LMS Algorithm Using the Error Difference — Research Paper | ScholarLens