Over-the-air federated learning with energy harvesting devices
File(s)
Author(s)
Aygun, Ozan
Kazemi, Mohammad
Gunduz, Deniz
Duman, Tolga M
Type
Conference Paper
Abstract
We consider federated edge learning among mobile devices that harvest the required energy from their surroundings, and share their updates with the parameter server (PS) through a shared wireless channel. In particular, we consider energy harvesting FL with over-the-air (OTA) aggregation, where the participating devices perform local computations and wireless transmission only when they have the required energy available, and transmit the local updates simultaneously over the same channel bandwidth. In order to prevent bias among the heterogeneous devices, we utilize a weighted averaging with respect to their latest energy arrivals and data cardinalities. We provide a convergence analysis and carry out numerical experiments with different energy arrival profiles, which show that the proposed scheme is robust against heterogeneous energy arrivals in error-free scenarios while having less than 10% performance loss for fading channels.
Date Issued
2023-01-11
Date Acceptance
2022-12-01
Citation
GLOBECOM 2022 - 2022 IEEE Global Communications Conference, 2023, pp.1942-1947
ISSN
2334-0983
Publisher
IEEE
Start Page
1942
End Page
1947
Journal / Book Title
GLOBECOM 2022 - 2022 IEEE Global Communications Conference
Copyright Statement
Copyright © 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Source
IEEE Global Communications Conference (GLOBECOM)
Subjects
CHANNEL
Computer Science
Computer Science, Information Systems
Computer Science, Theory & Methods
energy harvesting devices
Engineering
Engineering, Electrical & Electronic
Federated learning
machine learning
Science & Technology
Technology
Telecommunications
TRANSMISSION
wireless communications
Publication Status
Published
Start Date
2022-12-04
Finish Date
2022-12-08
Coverage Spatial
Rio de Janeiro, Brazil