Publication Details

A Novel Variable Step-Size LMS Algorithm for Decentralized Incremental Distributed Networks

QADRI Syed Safi Uddin, ARIF Muhammad and SAEED Bin Omer Muhammad. A Novel Variable Step-Size LMS Algorithm for Decentralized Incremental Distributed Networks. Circuits, Systems, and Signal Processing, vol. 41, no. 12, 2023, pp. 7226-7249. ISSN 0278-081X. Available from: https://link.springer.com/article/10.1007/s00034-023-02426-y
Czech title
Nový algoritmus LMS s proměnlivou velikostí kroku pro decentralizované inkrementální distribuované sítě
Type
journal article
Language
english
Authors
Qadri Syed Safi Uddin (KIET)
Arif Muhammad, Ph.D. (DCSY FIT BUT)
Saeed Bin Omer Muhammad ()
URL
Keywords

Variable step-size, Least mean square algorithm, Incremental algorithm, Mean squared deviation, Theoretical analysis, Quotient form

Abstract

This work proposes a variable step-size strategy for estimation over distributed networks using the incremental scheme. The proposed algorithm employs the ratio of filtered squared instantaneous error to the squared instantaneous error in a windowed format. This reduces the dependency on the error power which is particularly beneficial in low signal-to-noise power ratio situations. A comprehensive theoretical analysis has been performed, and closed-form solutions of mean squared error, excess mean squared error and mean squared deviation have been derived. The theoretical results are verified via simulation results. Extensive testing has been done through experiments under various scenarios to show the supremacy of the proposed algorithm in comparison with several other algorithms.

Published
2023
Pages
7226-7249
Journal
Circuits, Systems, and Signal Processing, vol. 41, no. 12, ISSN 0278-081X
Publisher
Springer Nature Switzerland AG
DOI
UT WoS
001034471300001
EID Scopus
BibTeX
@ARTICLE{FITPUB12873,
   author = "Uddin Safi Syed Qadri and Muhammad Arif and Muhammad Omer Bin Saeed",
   title = "A Novel Variable Step-Size LMS Algorithm for Decentralized Incremental Distributed Networks",
   pages = "7226--7249",
   journal = "Circuits, Systems, and Signal Processing",
   volume = 41,
   number = 12,
   year = 2023,
   ISSN = "0278-081X",
   doi = "10.1007/s00034-023-02426-y",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12873"
}
Back to top