Variable step size modified clipped LMS algorithm
Publish Year: 1394
نوع سند: مقاله کنفرانسی
زبان: English
View: 699
This Paper With 5 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
KBEI02_292
تاریخ نمایه سازی: 5 بهمن 1395
Abstract:
In this paper we introduce an Modified Clipped LMS (MCLMS) algorithm with a variable step size. In the MCLMS algorithm two parameters, the step size and the threshold control the convergence rate of the adaptive filter coefficients and also determine the final mean-square error. The computational complexity decreased dramatically by a large threshold. However, this selection results in a low convergence rate. Since the convergence time is inversely proportional to the step size, a large step size is often selected for fast convergence. But a large step size results in an increased final mean square error. Therefore in this paper we choose a large threshold and propose a variable step size for the MCLMS algorithm. The advantages of this proposed variable step size and a large threshold selection are that the computation complexity is low, final mean square error is low and that the convergence is fast.
Keywords:
adaptive filter , LMS algorithm , modified clipped LMS algorithm , variable step size , convergence rate
Authors
Amin Aref
Department of Electrical and Computer Eng. Shahrood University Shahrood, Iran
Mojtaba Lotfizad
Department of Electrical and Computer Eng. Tarbiat Modares University Tehran, Iran