面向6G网络的联邦学习高效训练优化方法

苏瑞; 朱亚丽; 朱晓荣

doi:null

您当前的位置：

首页 >

文章列表页 >

面向6G网络的联邦学习高效训练优化方法

理论与技术 | 更新时间：2026-03-03

- 面向6G网络的联邦学习高效训练优化方法
- An efficient training optimization method for federated learning in 6G networks
- 物联网学报 2026年
- 作者机构：
  
  1.南京邮电大学通信与信息工程学院，江苏南京 210003
  2.江苏商贸职业学院，江苏南通 226000
- 作者简介：
  
  朱晓荣，xrzhu@njupt.edu.cn
- 基金信息：
  
  国家自然科学基金资助项目(92367102);南通市科技计划项目(MS2023062)
- DOI：
  中图分类号： TN929.5
- 收稿：2024-12-03，
  
  修回：2025-01-10，
  
  录用：2025-02-21，
- 稿件说明：
移动端阅览
苏瑞, 朱亚丽, 朱晓荣. 面向6G网络的联邦学习高效训练优化方法[J/OL]. 物联网学报, 2026.

SU Rui, ZHU Yali, ZHU Xiaorong. An efficient training optimization method for federated learning in 6G networks[J/OL]. Chinese Journal on Internet of Things, 2026.
苏瑞, 朱亚丽, 朱晓荣. 面向6G网络的联邦学习高效训练优化方法[J/OL]. 物联网学报, 2026. DOI：

SU Rui, ZHU Yali, ZHU Xiaorong. An efficient training optimization method for federated learning in 6G networks[J/OL]. Chinese Journal on Internet of Things, 2026. DOI：

摘要

第六代移动通信技术(6G)预计将充分利用大量敏感数据并有望实现一个完全互联的世界，联邦学习作为一种新兴的分布式计算范式广泛应用于各领域。在传统的联邦学习方法中，收敛前的大量通信轮次导致了巨大的通信成本。此外，在每一轮训练中，客户端在上行通信阶段向参数服务器上传大量的模型参数，增加了通信开销。针对6G网络提出了一种高效的联邦学习训练优化方法(CEFED)，采用定制化的局部训练策略，通过约束项在更少的通信轮数内实现快速收敛。考虑到上行链路拥塞，引入了部分客户端参与规则，以避免多个客户端同时上传参数更新。同时针对数据异质性导致的局部模型之间的差异，提出了一种基于数据质量的加权聚合策略，通过为不同本地参数分配合理权值以提升模型精度。实验结果表明，该算法在多种联邦学习设置下，均超越了经典的FedAVG算法。与基准算法相比，CEFED算法不仅在提升测试精度的基础上加快了模型的收敛速度，而且有效节省了通信开销。

Abstract

The sixth generation of mobile communication technology (6G) is expected to make full use of a vast amount of sensitive data and is anticipated to achieve a completely interconnected world. Federated learning

as an emerging distributed computing paradigm

is widely applied in various fields. In traditional federated learning methods

a large number of communication rounds before convergence lead to significant communication costs. Moreover

during each training round

clients upload a large amount of model parameters to the parameter server during the uplink phase

increasing communication overhead. An efficient federated learning training optimization method (CEFED) has been proposed for 6G networks

which adopts a customized local training strategy to achieve rapid convergence within fewer communication rounds through a constraint term. Considering uplink congestion

a partial client participation rule is introduced to prevent multiple clients from uploading parameter updates simultaneously. At the same time

in response to the differences between local models caused by data heterogeneity

a weighted aggregation strategy based on data value is proposed

which assigns reasonable weights to different local parameters to improve model accuracy. Experimental results show that the algorithm outperforms the classic FedAVG algorithm in various federated learning settings. Compared with the benchmark algorithm

the CEFED algorithm not only accelerates the model's convergence rate while improving test accuracy

but also effectively saves on communication costs.

关键词

Keywords

references

耿光磊 , 高博 , 熊轲 , 等 . 联邦学习赋能6G网络综述 [J ] . 物联网学报 , 2023 , 7 ( 2 ): 50 - 66 .

GENG G L , GAO B , XIONG K , et al . A survey of federated learning for 6G networks [J ] . Chinese Journal on Internet of Things , 2023 , 7 ( 2 ): 50 - 66 .

李文璟 , 喻鹏 , 张平 . 6G智能内生网络架构及关键技术分析 [J ] . 中兴通讯技术 , 2023 , 29 ( 05 ): 2 - 8 .

LI W J , YU P , ZHANG P . Analysis of the architecture and key technologies of 6G intelligent endogenous networks [J ] . ZTE Technology Journal , 2023 , 29 ( 05 ): 2 - 8 .

ZHOU F Q , LI W J , YANG Y , et al . Intelligence-endogenous networks: Innovative network paradigm for 6G [J ] . IEEE Wireless Communications , 2022 , 29 ( 1 ): 40 - 47 .

栾宁 , 熊轲 , 张煜 , 等 . 6G: 典型应用, 关键技术与面临挑战 [J ] . 物联网学报 , 2022 , 6 ( 1 ): 29 - 43 .

LUAN N , XIONG K , ZHANG Y , et al . 6G: typical applications, key technologies and challenges [J ] . Chinese Journal on Internet of Things , 2022 , 6 ( 1 ): 29 - 43 .

MCMAHAN H B , MOORE E , RAMAGE D , et al . Communication-efficient learning of deep networks from decentralized data [C ] // Artificial intelligence and statistics . PMLR , 2017 : 1273 - 1282 .

GHADIMI S , LAN G H . Stochastic first-and zeroth-order methods for nonconvex stochastic programming [J ] . SIAM journal on optimization , 2013 , 23 ( 4 ): 2341 - 2368 .

PENG Y Q , HAO Z Y , YUN X C . Lock-free parallelization for variance-reduced stochastic gradient descent on streaming data [J ] . IEEE Transactions on Parallel and Distributed Systems , 2020 , 31 ( 9 ): 2220 - 2231 .

ZHANG J , GUO S , QU Z H , et al . Adaptive federated learning on non-iid data with resource constraint [J ] . IEEE Transactions on Computers , 2021 , 71 ( 7 ): 1655 - 1667 .

王勇 , 李国良 , 李开宇 . 联邦学习贡献评估综述 . 软件学报 , 2023 , 34 ( 3 ): 1168 - 1192 .

WANG Y , LI G L , LI K Y . Survey on contribution evaluation for federated learning [J ] . Journal of Software , 2023 , 34 ( 3 ): 1168 - 1192 .

王波 , 代晓蕊 , 王伟 , 等 . 面向联邦学习的对抗样本投毒攻击 [J ] . 中国科学: 信息科学 , 2023 , 53 ( 3 ): 470 - 484 .

WANG B , DAI X R , WANG W , et al . Adversarial examples for poisoning attacks against federated learning [J ] . Science in China(Information Sciences) , 2023 , 53 ( 3 ): 470 - 484 .

SOLTANI B , HAGHIGHI V , MAHMOOD A , et al . A survey on participant selection for federated learning in mobile networks [C ] // Proceedings of the 17th ACM Workshop on Mobility in the Evolving Internet Architecture . 2022 : 19 - 24 .

GUPTA A , LUO T , NGO M V , et al . Long-short history of gradients is all you need: Detecting malicious and unreliable clients in federated learning [C ] // European Symposium on Research in Computer Security . Cham : Springer Nature Switzerland , 2022 : 445 - 465 .

WU J J , DONG F , LEUNG H , et al . Topology-aware federated learning in edge computing: A comprehensive survey [J ] . ACM Computing Surveys , 2024 , 56 ( 10 ): 1 - 41 .

LI T , SAHU A K , ZAHEER M , et al . Federated optimization in heterogeneous networks [J ] . Proceedings of Machine learning and systems , 2020 , 2 : 429 - 450 .

LI Q B , HE B S , SONG D . Model-contrastive federated learning [C ] // Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 2021 : 10713 - 10722 .

KARIMIREDDY S P , KALE S , MOHRI M , et al . Scaffold: Stochastic controlled averaging for federated learning [C ] // International conference on machine learning . PMLR , 2020 : 5132 - 5143 .

FALLAH A , MOKHTARI A , OZDAGLAR A . Personalized federated learning with theoretical guarantees: A model-agnostic meta-learning approach [J ] . Advances in neural information processing systems , 2020 , 33 : 3557 - 3568 .

CHO Y J , WANG J Y , JOSHI G . Towards understanding biased client selection in federated learning [C ] // International Conference on Artificial Intelligence and Statistics . PMLR , 2022 : 10351 - 10375 .

WANG H , KAPLAN Z , NIU D , et al . Optimizing federated learning on non-iid data with reinforcement learning [C ] // IEEE INFOCOM 2020-IEEE conference on computer communications . IEEE , 2020 : 1698 - 1707 .

NISHIO T , YONETANI R . Client selection for federated learning with heterogeneous resources in mobile edge [C ] // ICC 2019-2019 IEEE international conference on communications (ICC) . IEEE , 2019 : 1 - 7 .

CHAI Z , ALI A , ZAWAD S , et al . Tifl: A tier-based federated learning system [C ] // Proceedings of the 29th international symposium on high-performance parallel and distributed computing . 2020 : 125 - 136 .

LAI F , ZHU X F , MADHYASTHA H V , et al . Oort: Efficient federated learning via guided participant selection [C ] // Proceedings of the 15th USENIX Symposium on Operating Systems Design and Implementation , 2021 : 19 - 35 .

CHAI Z , CHEN Y J , ANWAR A , et al . FedAT: A high-performance and communication-efficient federated learning system with asynchronous tiers [C ] // Proceedings of the international conference for high performance computing, networking, storage and analysis . 2021 : 1 - 16 .

WU H D , WANG P . Fast-convergent federated learning with adaptive weighting [J ] . IEEE Transactions on Cognitive Communications and Networking , 2021 , 7 ( 4 ): 1078 - 1088 .

AAMER B , CHERGUI H , BENJILLALI M , et al . Entropy-driven stochastic policy for fast federated learning in beyond 5G edge-RAN [C ] // 2021 IEEE Global Communications Conference (GLOBECOM) . IEEE , 2021 : 1 - 6 .

WANG D , ZHANG N F , TAO M X . Clustered federated learning with weighted model aggregation for imbalanced data [J ] . China Communications , 2022 , 19 ( 8 ): 41 - 56 .

ZHAO B R , CUI Q M , LIANG S Y , et al . Green concerns in federated learning over 6G [J ] . China Communicaitons , 2022 , 19 ( 3 ): 50 - 69 .

YU A , YANG Q K , DOU L H , et al . Federated imitation learning: a cross-domain knowledge sharing framework for traffic scheduling in 6G ubiquitous IoT [J ] . IEEE Network , 2021 , 35 ( 5 ): 136 - 142 .

ZHU Z Q , Wan S , Fan P Y , et al . Federated multiagent actor–critic learning for age sensitive mobile-edge computing [J ] . IEEE Internet of Things Journal , 2021 , 9 ( 2 ): 1053 - 1067 .

LU Z L , PAN H , DAI Y Y , et al . Federated learning with non-iid data: A survey [J ] . IEEE Internet of Things Journal , 2024 , 11 ( 11 ), 19188 - 19209 .

ZENG Y , MU Y K , YUAN J F , et al . Adaptive federated learning with non-IID data [J ] . The Computer Journal , 2023 , 66 ( 11 ): 2758 - 2772 .

YAN Z , YI Y Z , ZHANG J L , et al . Federated learning model training method based on data features perception aggregation [C ] // 2021 IEEE 94th Vehicular Technology Conference (VTC2021-Fall) . IEEE , 2021 : 1 - 7 .

XU X Y , LYU L J , MA X J , et al . Gradient driven rewards to guarantee fairness in collaborative machine learning [J ] . Advances in Neural Information Processing Systems , 2021 , 34 : 16104 - 16117 .

XU X Y , WU Z X , FOO C S , et al . Validation free and replication robust volume-based data valuation [J ] . Advances in Neural Information Processing Systems , 2021 , 34 : 10837 - 10848 .

MA X D , ZHU J , LIN Z H , et al . A state-of-the-art survey on solving non-iid data in federated learning [J ] . Future Generation Computer Systems , 2022 , 135 : 244 - 258 .

KRIZHEVSKY A , SUTSKEVER I , HINTON G E . Imagenet classification with deep convolutional neural networks [J ] . Advances in neural information processing systems , 2012 , 25 .

FAN X , WANG Y , HUO Y , et al . Bev-sgd: Best effort voting sgd against byzantine attacks for analog-aggregation-based federated learning over the air [J ] . IEEE Internet of Things Journal , 2022 , 9 ( 19 ): 18946 - 18959 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向无线联邦学习模型压缩的多维资源联合优化研究

车联网中联邦学习模型低时延传输迁移方法研究

面向工业物联网的时延感知半同步联邦学习客户端资源联合调度方案

面向联邦学习标签翻转攻击的客户端选择防御方法