An optimized ZA-LMS algorithm for time varying sparse system

The zero attracting least mean square algorithm has improved performance than conventional LMS when the system is sparse and its performance decreases when the sparsity level is decreased or when the system is time varying. The proposed algorithm focused on optimization of both step size and zero at...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of speech technology Vol. 22; no. 2; pp. 441 - 447
Main Authors Radhika, S., Arumugam, Chandrasekar
Format Journal Article
LanguageEnglish
Published New York Springer US 15.06.2019
Springer Nature B.V
Subjects
Algorithms
Artificial Intelligence
Computer simulation
Convergence
Engineering
Optimization
Signal,Image and Speech Processing
Simulation
Social Sciences
Sparsity
State variable
System identification
Time
Step size
Sparsity
State variable model
Zero attractor controller
Convergence
Online AccessGet full text
ISSN1381-2416
1572-8110
DOI10.1007/s10772-019-09616-7

Cover

More Information
Summary:The zero attracting least mean square algorithm has improved performance than conventional LMS when the system is sparse and its performance decreases when the sparsity level is decreased or when the system is time varying. The proposed algorithm focused on optimization of both step size and zero attractor controller using state variable model to improve the overall performance at all sparsity levels. Simulations in the context of time varying sparse system identification proved that the proposed algorithm provides good performance when compared to the conventional ones.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1381-2416
1572-8110
DOI:10.1007/s10772-019-09616-7