Efficient Time-Varying q-Parameter Design for q-Incremental Least Mean Square Algorithm with Noisy Links

Muhammad Arif, Saba Samreen Khan, Syed Safi Uddin Qadri, Imran Naseem, Muhammad Moinuddin

Research output: Contribution to journalArticlepeer-review

Abstract

The quantum calculus provides an extra degree of freedom to search the local and global minima by inducing a q-parameter. Motivated by this fact, a quantum calculus-based noisy links incremental least mean squares (NL-qILMS) algorithm is proposed. Moreover, for the proposed NL-qILMS, we also devised various time-varying techniques for the selection of the optimal q-parameter to improve the performance. Furthermore, the closed-form solutions for the steady-state mean square deviation, excess mean square deviation and mean square error are derived. The analytical results are validated through simulation. Finally, extensive simulations have been done to evaluate the performance of the proposed algorithm for various choices of optimal q-values.

Original languageEnglish
Pages (from-to)5699-5718
Number of pages20
JournalCircuits, Systems, and Signal Processing
Volume41
Issue number10
Early online date2022
DOIs
Publication statusPublished - Oct 2022

Fingerprint

Dive into the research topics of 'Efficient Time-Varying q-Parameter Design for q-Incremental Least Mean Square Algorithm with Noisy Links'. Together they form a unique fingerprint.

Cite this