A new efficient LMS adaptive filtering algorithm

A new LMS adaptive filtering algorithm is proposed. The algorithm has comparable performance to the direct-form LMS algorithm (DLMS), while costs N/2-1 less multiplications at the expense of N/2+3 more additions than DLMS algorithm. For coefficient update, the new algorithm needs one additional coef...

Full description

Saved in:
Bibliographic Details
Main Authors: Sau-Gee Chen, Yung-An Kao, Kwung-Yee Tsai
Format: Conference Proceeding
Language:English
Subjects:
Adaptive algorithm
Adaptive filters
Algorithm design and analysis
Analytical models
Convolution
Costs
Counting circuits
Filtering algorithms
Least squares approximation
Performance analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A new LMS adaptive filtering algorithm is proposed. The algorithm has comparable performance to the direct-form LMS algorithm (DLMS), while costs N/2-1 less multiplications at the expense of N/2+3 more additions than DLMS algorithm. For coefficient update, the new algorithm needs one additional coefficient estimation. Further, the algorithm is combined with LMS sign algorithm (SA), signed regressor algorithm (SRA) and zero forcing (ZFA) algorithm for further complexity reduction. Simulation results confirm with theoretical analysis that the new algorithm and its SA, SRA and ZFA versions converge as fast as their counter DLMS algorithms, while maintaining comparable performance.< >
DOI:10.1109/TENCON.1994.369225