Show simple item record

AuthorHamdi, Rami
AuthorChen, Mingzhe
AuthorBen Said, Ahmed
AuthorQaraqe, Marwa
AuthorPoor, H. Vincent
Available date2023-10-08T08:41:46Z
Publication Date2022
Publication NameIEEE Internet of Things Journal
ResourceScopus
ISSN23274662
URIhttp://dx.doi.org/10.1109/JIOT.2021.3089054
URIhttp://hdl.handle.net/10576/48315
AbstractIn this article, the deployment of federated learning (FL) is investigated in an energy harvesting wireless network in which the base stations (BSs) employs massive multiple-input-multiple-output (MIMO) to serve a set of users powered by independent energy harvesting sources. Since a certain number of users may not be able to participate in FL due to interference and energy constraints, a joint energy management and user scheduling problem in FL over wireless systems is formulated. This problem is formulated as an optimization problem whose goal is to minimize the FL training loss via optimizing user scheduling. To find how the transmit power, the number of scheduled users and user association, affect the training loss, the FL convergence rate is first analyzed. Given this analytical result, the original optimization problem can be decomposed, simplified, and solved. Simulation results show that the proposed user scheduling and user association algorithm can reduce training loss compared to a standard FL algorithm.
SponsorThis work was supported in part by the TÜBiTAK-QNRF Joint Funding Program from the Scientific and Technological Research Council of Turkey and Qatar National Research Fund (QNRF) under Grant AICC03-0324-200005, and in part by the U.S. National Science Foundation under Grant CCF-1908308.
Languageen
PublisherInstitute of Electrical and Electronics Engineers Inc.
SubjectEnergy harvesting
federated learning (FL)
resource allocation
TitleFederated Learning over Energy Harvesting Wireless Networks
TypeArticle
Pagination92-103
Issue Number1
Volume Number9
dc.accessType Abstract Only


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record