Analysis of an improved variable order least mean square algorithm

被引：0

作者：

Zhang Y.-G. ^{[1
]}

Li N. ^{[1
]}

Hao Y.-L. ^{[1
]}

机构：

[1] College of Automation, Harbin Engineering University

来源：

Harbin Gongcheng Daxue Xuebao/Journal of Harbin Engineering University | 2010年 / 31卷 / 03期

关键词：

Adaptive algorithm; Adaptive filter; Least mean square; Variable error width; Variable order;

D O I：

10.3969/j.issn.1006-7043.2010.03.014

中图分类号：

学科分类号：

摘要：

Based on analysis of the fractional variable order least mean square (LMS) algorithm, an improved variable order LMS algorithm was proposed. Analysis of the variable order LMS algorithm showed that to reduce its steady state order error, a variable error width parameter must be added to the proposed algorithm. Theoretical analysis provided parameter choice guideline for the proposed algorithm. Simulations were performed under low noise and high noise conditions, and also on a system with sparse distribution of impulse response (curve). The results showed that, by comparing with the variable order LMS algorithm respectively under the above three simulation conditions, when the coefficient of convergence was the same, the steady state order error of the proposed algorithm reduced to 10%; when the SNR was 0dB, the steady state order error reduced to 1/3.

引用

页码：350 / 354

页数：4

共 9 条

[1]

Farhang-Boroujeny B., Adaptive Filters: Theory And Applications, pp. 1-24, (1998)

[2]

Sayed A.H., Fundamentals of Adaptive Filtering, pp. 1-12, (2003)

[3]

Gu Y., Tang K., Cui H., Et al., Convergence analysis of a deficient-length LMS filter and optimal-length sequence to model exponential decay impulse response, IEEE Signal Process Letters, 10, 1, pp. 4-7, (2003)

[4]

Mayyas K., Performance analysis of the deficient length LMS adaptive algorithm, IEEE Tran Signal Processing, 53, 8, pp. 2727-2734, (2005)

[5]

Riero-Palou F., Noras J.M., Cruickshank D.J.M., Linear equalisers with dynamic and automatic length selection, Electron Letters, 37, 25, pp. 1553-1554, (2001)

[6]

Gu Y., Tang K., Cui H., Et al., LMS algorithm with gradient descent filter length, IEEE Signal Process Letters, 11, 3, pp. 305-307, (2004)

[7]

Gong Y., Cowan C.F.N., An LMS style variable tap-length algorithm for structure adaptation, IEEE Tran Signal Processing, 53, 7, pp. 2400-2407, (2005)

[8]

Zhang Y., Li N., Chambers J.A., Et al., Steady-state performance analysis of a variable tap-length LMS algorithm, IEEE Tran Signal Processing, 56, 2, pp. 839-845, (2008)

[9]

Greenberg J.E., Modified LMS algorithm for speech processing with an adaptive noise canceller, IEEE Trans Signal Processing, 6, 4, pp. 338-351, (1998)

← 1 →