Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness

Abstract With the explosive growth of smart mobile devices in wireless networks, the increasing computational power of mobile chips and the growing concern for personal privacy, a decentralized deep learning framework at the mobile terminal layer has emerged called federated learning (FL) to enhance...

Full description

Saved in:
Bibliographic Details
Main Authors: Guohao Pang, Xiaorong Zhu
Format: Article
Language:English
Published: SpringerOpen 2024-11-01
Series:EURASIP Journal on Advances in Signal Processing
Subjects:
Online Access:https://doi.org/10.1186/s13634-024-01192-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850064367270232064
author Guohao Pang
Xiaorong Zhu
author_facet Guohao Pang
Xiaorong Zhu
author_sort Guohao Pang
collection DOAJ
description Abstract With the explosive growth of smart mobile devices in wireless networks, the increasing computational power of mobile chips and the growing concern for personal privacy, a decentralized deep learning framework at the mobile terminal layer has emerged called federated learning (FL) to enhance user experience. This paper studies the training efficiency optimization problem of wireless FL that jointly considers processor performance, channel conditions and terminals’ power in a non-independent identically distribution (non-IID) scenario. And, the training efficiency optimization problem is mathematically modeled and then decomposed into several sub-problems based on the independence and decoupling of the variables involved. To enhance the training efficiency of wireless FL, a comprehensive scheduling strategy encompassing computational and communication aspects is proposed. Simulation results show that the proposed scheduling strategy for wireless FL achieves superior learning performance with reduced training latency.
format Article
id doaj-art-507a1398a0e149348ff06d2f71e0669d
institution DOAJ
issn 1687-6180
language English
publishDate 2024-11-01
publisher SpringerOpen
record_format Article
series EURASIP Journal on Advances in Signal Processing
spelling doaj-art-507a1398a0e149348ff06d2f71e0669d2025-08-20T02:49:19ZengSpringerOpenEURASIP Journal on Advances in Signal Processing1687-61802024-11-012024112810.1186/s13634-024-01192-6Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awarenessGuohao Pang0Xiaorong Zhu1The College of Portland, Nanjing University of Posts and TelecommunicationsThe College of Telecommunications and Information Engineering, Nanjing University of Posts and TelecommunicationsAbstract With the explosive growth of smart mobile devices in wireless networks, the increasing computational power of mobile chips and the growing concern for personal privacy, a decentralized deep learning framework at the mobile terminal layer has emerged called federated learning (FL) to enhance user experience. This paper studies the training efficiency optimization problem of wireless FL that jointly considers processor performance, channel conditions and terminals’ power in a non-independent identically distribution (non-IID) scenario. And, the training efficiency optimization problem is mathematically modeled and then decomposed into several sub-problems based on the independence and decoupling of the variables involved. To enhance the training efficiency of wireless FL, a comprehensive scheduling strategy encompassing computational and communication aspects is proposed. Simulation results show that the proposed scheduling strategy for wireless FL achieves superior learning performance with reduced training latency.https://doi.org/10.1186/s13634-024-01192-6Federated learningWireless communicationScheduling policiesParallel and distributed algorithmsResource allocation
spellingShingle Guohao Pang
Xiaorong Zhu
Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
EURASIP Journal on Advances in Signal Processing
Federated learning
Wireless communication
Scheduling policies
Parallel and distributed algorithms
Resource allocation
title Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
title_full Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
title_fullStr Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
title_full_unstemmed Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
title_short Training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
title_sort training efficiency optimization algorithm of wireless federated learning based on processor performance and network condition awareness
topic Federated learning
Wireless communication
Scheduling policies
Parallel and distributed algorithms
Resource allocation
url https://doi.org/10.1186/s13634-024-01192-6
work_keys_str_mv AT guohaopang trainingefficiencyoptimizationalgorithmofwirelessfederatedlearningbasedonprocessorperformanceandnetworkconditionawareness
AT xiaorongzhu trainingefficiencyoptimizationalgorithmofwirelessfederatedlearningbasedonprocessorperformanceandnetworkconditionawareness