基于演化博弈与中断概率优化的半分布式联邦边缘学习激励机制

刘雨欣; 王一航; 王思洋; 夏文超; 赵海涛; 卜宪德

doi:10.12263/DZXB.20250658

您当前的位置：

首页 >

文章列表页 >

基于演化博弈与中断概率优化的半分布式联邦边缘学习激励机制

学术论文 | 更新时间：2026-02-10

- 基于演化博弈与中断概率优化的半分布式联邦边缘学习激励机制
- Incentive Mechanism Based on Evolutionary Game and Interruption Probability Optimization for Semi-Distributed Federated Edge Learning
- 电子学报 2025年53卷第11期页码：3956-3969
- 作者机构：
  
  1.南京邮电大学教育部泛在网络健康服务系统工程研究中心，江苏南京 210003
  2.网络通信与安全紫金山实验室，江苏南京 211111
  3.中国电力科学研究院有限公司信息通信研究所，江苏南京 210003
- 作者简介：
  
  [ "刘雨欣女，2003年6月出生于四川省德阳市.现为南京邮电大学通信与信息工程学院本科生.主要研究方向为联邦学习激励机制.E-mail: b22010703@njupt.edu.cn" ]
  [ "王一航男，2004年3月出生于江苏省苏州市.现为南京邮电大学本科生.主要研究方向为通信工程.E-mail: p22000407@njupt.edu.cn" ]
  [ "王思洋女，2000年2月出生于辽宁省沈阳市.现为南京邮电大学通信与信息工程学院博士研究生.主要研究方向为联邦学习.E-mail: 2024010101@njupt.edu.cn" ]
  [ "夏文超男，1991年4月出生于江苏省盐城市.现为南京邮电大学通信与信息工程学院副教授.主要研究方向为通感算协同技术、群智协同技术等.E-mail: xiawenchao@njupt.edu.cn" ]
  [ "赵海涛男，1983年10月出生于山东省潍坊市.现为南京邮电大学物联网学院教授.主要研究方向为智能网络技术、多信道建模技术、物联网、边缘计算等.中国电子学会会员编号：E190022504S.E-mail: zhaoht@njupt.edu.cn" ]
  [ "卜宪德男，1978年10月出生于江苏省徐州市.现为中国电力科学研究院有限公司正高级工程师.主要研究方向为电力无线接入网络、异构无线网络通信技术等.E-mail: buxiande2@epri.sgcc.com.cn" ]
- 基金信息：
  
  江苏省基础研究计划（自然科学基金）前沿引领技术基础研究专项(BK20212001);青年人才托举工程(2022QNRC001)
- DOI：10.12263/DZXB.20250658
  中图分类号： TP181;
- 收稿：2025-07-28，
  
  录用：2025-11-10，
  
  纸质出版：2025-11-25
- 稿件说明：
移动端阅览
刘雨欣, 王一航, 王思洋, 等. 基于演化博弈与中断概率优化的半分布式联邦边缘学习激励机制[J]. 电子学报, 2025, 53(11): 3956-3969.

LIU Yu-xin, WANG Yi-hang, WANG Si-yang, et al. Incentive Mechanism Based on Evolutionary Game and Interruption Probability Optimization for Semi-Distributed Federated Edge Learning[J]. Acta Electronica Sinica, 2025, 53(11): 3956-3969.
刘雨欣, 王一航, 王思洋, 等. 基于演化博弈与中断概率优化的半分布式联邦边缘学习激励机制[J]. 电子学报, 2025, 53(11): 3956-3969. DOI：10.12263/DZXB.20250658

LIU Yu-xin, WANG Yi-hang, WANG Si-yang, et al. Incentive Mechanism Based on Evolutionary Game and Interruption Probability Optimization for Semi-Distributed Federated Edge Learning[J]. Acta Electronica Sinica, 2025, 53(11): 3956-3969. DOI：10.12263/DZXB.20250658

摘要

随着物联网快速发展，大规模泛在分布终端产生的数据规模激增.为提升网络智慧服务能力，采用半分布式联邦边缘学习（Semi-Decentralized FEderated Edge Learning，SD-FEEL）方式，通过多个边缘服务器各自协调一个终端簇执行本地更新，边缘服务器之间周期性共享模型更新，可以在保障学习性能的同时有效缓解网络拥塞.然而，在实际部署中，缺乏有效激励会导致终端参与学习的积极性不高，且训练过程中可能发生无线通信中断，这些因素都会降低模型训练效率.因此，本文提出一种面向SD-FEEL场景的基于演化博弈与中断概率优化的激励机制.具体来说，首先，设计同时考虑数据大小与质量的终端贡献评估指标，并据此构建奖励函数激励高质量终端参与训练，提升全局模型性能并确保激励公平性.其次，引入演化博弈框架，捕捉终端的有限理性和动态决策行为，均衡边缘服务器负载，确定种群中关联各边缘服务器的最优比例，实现种群总效益最大化.在此基础上，以最小化无线通信中断概率为目标，优化终端与边缘服务器的具体关联策略.仿真结果表明，所提方法能够有效均衡边缘服务负载，与随机接入方案和声誉激励机制（Reputation-Aware Incentive Mechanism，RAIM）方案相比，通信中断概率分别降低32.04%和35.55%，模型准确性分别提升3.58%和4.34%.

Abstract

With the rapid advancement of the Internet of Things

the volume of data generated by large-scale

ubiquitous distributed terminals is surging. To enhance the intelligence of network services

we adopt the semi-decentralized federated edge learning (SD-FEEL) paradigm

in which multiple edge servers each coordinate a cluster of terminals to perform local model updates

while periodically exchanging updates among themselves. This approach preserves learning performance while effectively mitigating network congestion. However

real-world deployments encounter key challenges: inadequate incentives reduce terminals’ motivation to participate in training

and wireless communication interruptions during the process can degrade overall training efficiency. To address these issues

this paper proposes an incentive mechanism for SD-FEEL scenarios

leveraging evolutionary game theory and optimization of interruption probabilities. Specifically

first

we design a terminal contribution metric that incorporates both data quantity and quality

along with a corresponding reward function to encourage participation from high-quality terminals. This not only boosts global model performance but also ensures fairness in incentives. Second

we introduce an evolutionary game framework to model terminals’ bounded rationality and dynamic decision-making behaviors. This framework balances edge server loads

determines the optimal proportions of terminals associating with each server within the population

and maximizes the population’s overall utility. Building on this foundation

we further optimize specific terminal-to-edge server association strategies with the goal of minimizing the probability of wireless communication interruptions. Simulation results demonstrate that the proposed method can effectively balance the edge service load. Compared to the random access method and the reputation-aware incentive mechanism(RAIM) scheme

the communication interruption probability is reduced by 32.04% and 35.55% respectively

and the model accuracy is improved by 3.58% and 4.34%

respectively.

关键词

Keywords

references

SCHILLER E , AIDOO A , FUHRER J , et al . Landscape of IoT security [J ] . Computer Science Review , 2022 , 44 ( 1 ): 1 - 18 .

HUSEN A , CHAUDARY M H , AHMAD F , et al . Holistic learning and optimization scheme for wireless networks [J ] . Transactions on Emerging Telecommunications Technologies , 2024 , 35 ( 10 ): e5036 .

DUTTA S . AI and IoT-based smart healthcare solutions in urban area [J ] . Trends in Health Informatics , 2025 , 2 ( 1 ): 18 - 26 .

IGE A B , ADEPOJU P A , AKINADE A O , et al . Machine learning in industrial applications: An in-depth review and future directions [J ] . International Journal of Multidisciplinary Research and Growth Evaluation , 2025 , 6 ( 1 ): 36 - 44 .

TIWARI P . The machine learning framework for traffic management in smart cities [J ] . Management of Environmental Quality , 2024 , 35 ( 2 ): 445 - 462 .

王鑫 , 黄伟口 , 孙凌云 . 跨机构联邦学习的激励机制综述 [J ] . 计算机科学 , 2024 , 51 ( 3 ): 20 - 29 .

WANG X , HUANG W K , SUN L Y . Survey of incentive mechanism for cross-silo federated learning [J ] . Computer Science , 2024 , 51 ( 3 ): 20 - 29 . (in Chinese)

陈颢瑜 , 李浥东 , 张洪磊 , 等 . 面向可信联邦学习公平性的研究综述 [J ] . 电子学报 , 2023 , 51 ( 10 ): 2985 - 3010 .

CHEN H Y , LI Y D , ZHANG H L , et al . Fairness in trustworthy federated learning: A survey [J ] . Acta Electronica Sinica , 2023 , 51 ( 10 ): 2985 - 3010 . (in Chinese)

RUAN M Z , LI Y H , ZHANG W Z , et al . Optimal power control for over-the-air federated learning with gradient compression [C ] // 2024 IEEE 30th International Conference on Parallel and Distributed Systems . Piscataway : IEEE , 2024 : 326 - 333 .

MCMAHAN H B , MOORE E , RAMAGE D , et al . Communication-efficient learning of deep networks from decentralized data [C ] // Proceedings of the 20th International Conference on Artificial Intelligence and Statistics . Cambridge : PMLR , 2017 ( 54 ): 1273 - 1282 .

马文玉 , 陈谦 , 胡宇翔 , 等 . 面向联邦算力物联网的隐私预算自适应优化方案 [J ] . 物联网学报 , 2024 , 8 ( 4 ): 54 - 69 .

MA W Y , CHEN Q , HU Y X , et al . A privacy budget adaptive optimization scheme for federated computing power Internet of Things [J ] . Chinese Journal on Internet of Things , 2024 , 8 ( 4 ): 54 - 69 . (in Chinese)

SUN Y C , SHAO J W , MAO Y Y , et al . Semi-decentralized federated edge learning for fast convergence on non-IID data [C ] // 2022 IEEE Wireless Communications and Networking Conference . Piscataway : IEEE , 2022 : 1898 - 1903 .

TIAN M M , CHEN Y X , LIU Y , et al . A contract theory based incentive mechanism for federated learning [EB/OL ] . ( 2021-08-12 )[ 2025-09-20 ] . https://arXiv.org/abs/2108.05568 https://arXiv.org/abs/2108.05568 .

LIM W Y B , HUANG J Q , XIONG Z H , et al . Towards federated learning in UAV-enabled Internet of vehicles: A multi-dimensional contract-matching approach [J ] . IEEE Transactions on Intelligent Transportation Systems , 2021 , 22 ( 8 ): 5140 - 5154 .

JIAO Y T , WANG P , NIYATO D , et al . Toward an automated auction framework for wireless federated learning services market [J ] . IEEE Transactions on Mobile Computing , 2021 , 20 ( 10 ): 3034 - 3048 .

杜辉 , 李卓 , 陈昕 . 基于在线双边拍卖的分层联邦学习激励机制 [J ] . 计算机科学 , 2022 , 49 ( 3 ): 23 - 30 .

DU H , LI Z , CHEN X . Incentive mechanism for hierarchical federated learning based on online double auction [J ] . Computer Science , 2022 , 49 ( 3 ): 23 - 30 . (in Chinese)

ZHANG J W , WU Y Z , PAN R . Auction-based ex-post-payment incentive mechanism design for horizontal federated learning with reputation and contribution measurement [EB/OL ] . ( 2022-03-15 )[ 2025-09-20 ] . https://arXiv.org/abs/2201.02410 https://arXiv.org/abs/2201.02410 .

LI C L , SONG M Y , LUO Y L . Federated learning based on Stackelberg game in unmanned-aerial-vehicle-enabled mobile edge computing [J ] . Expert Systems with Applications , 2024 , 235 : 121023 .

DANG Q L , YANG S , LIU Q Q , et al . Adaptive and communication-efficient zeroth-order optimization for distributed Internet of Things [J ] . IEEE Internet of Things Journal , 2024 , 11 ( 22 ): 37200 - 37213 .

WEIBULL J W . Evolutionary Game Theory [M ] . Cambridge : MIT Press , 1997 .

ZUO C H , XU P H , SONG Y C , et al . RAIM: Three-stage Stackelberg game for hierarchical federated learning with reputation-aware incentive mechanism [J ] . Scientific Reports , 2025 , 15 : 34344 .

康海燕 , 冀珊珊 . 面向无线边缘网络的分层Stackelberg博弈群体激励方法 [J ] . 电子学报 , 2024 , 52 ( 7 ): 2382 - 2392 .

KANG H Y , JI S S . Hierarchical Stackelberg game swarm learning incentive method for wireless edge network [J ] . Acta Electronica Sinica , 2024 , 52 ( 7 ): 2382 - 2392 . (in Chinese)

LIM W Y B , LUONG N C , HOANG D T , et al . Federated learning in mobile edge networks: A comprehensive survey [J ] . IEEE Communications Surveys & Tutorials , 2020 , 22 ( 3 ): 2031 - 2063 .

田有亮 , 吴柿红 , 李沓 , 等 . 基于激励机制的联邦学习优化算法 [J ] . 通信学报 , 2023 , 44 ( 5 ): 169 - 180 .

TIAN Y L , WU S H , LI T , et al . Federated learning optimization algorithm based on incentive mechanism [J ] . Journal on Communications , 2023 , 44 ( 5 ): 169 - 180 . (in Chinese)

LIM W Y B , NG J S , XIONG Z H , et al . Decentralized edge intelligence: A dynamic resource allocation framework for hierarchical federated learning [J ] . IEEE Transactions on Parallel and Distributed Systems , 2022 , 33 ( 3 ): 536 - 550 .

GOLDSMITH A . Wireless Communications [M ] . Cambridge : Cambridge University Press , 2005 .

ZHANG Z , JIANG J , XU H Y , et al . Distributed dynamic task allocation for unmanned aerial vehicle swarm systems: A networked evolutionary game-theoretic approach [J ] . Chinese Journal of Aeronautics , 2024 , 37 ( 6 ): 182 - 204 .

HOFBAUER J , SIGMUND K . Evolutionary game dynamics [J ] . Bulletin of the American Mathematical Society , 2003 , 40 ( 4 ): 479 - 519 .

IMPAGLIAZZO R , LOVETT S , PATURI R , et al . 0 - 1 integer linear programming with a linear number of constraints [EB/OL ] . ( 2014-02-18 )[ 2025-10-14 ] . https://arXiv.org/abs/1401.5512 https://arXiv.org/abs/1401.5512 .

SAM M L , SAPTARI A , SALLEH M R B , et al . Comparison between linear programming and integer linear programming: A review [J ] . International Journal of Mathematics in Operational Research , 2018 , 13 ( 1 ): 91 .

DEB K , PAL K . Efficiently solving: A large-scale integer linear program using a customized genetic algorithm [C ] // Genetic and Evolutionary Computation-GECCO 2004 . Berlin : Springer , 2004 : 1054 - 1065 .

耿方兴 , 李卓 , 陈昕 . 基于多领导者Stackelberg博弈的分层联邦学习激励机制设计 [J ] . 计算机应用 , 2023 , 43 ( 11 ): 3551 - 3558 .

GENG F X , LI Z , CHEN X . Incentive mechanism design for hierarchical federated learning based on multi-leader Stackelberg game [J ] . Journal of Computer Applications , 2023 , 43 ( 11 ): 3551 - 3558 . (in Chinese)

KRIZHEVSKY A . Learning multiple layers of features from tiny images [EB/OL ] . ( 2009-04-08 )[ 2025-09-10 ] . http://www.cs.utoronto.ca/~kriz/learning-features-2009-TR.pdf http://www.cs.utoronto.ca/~kriz/learning-features-2009-TR.pdf .

LECUN Y , BOTTOU L , BENGIO Y , et al . Gradient-based learning applied to document recognition [J ] . Proceedings of the IEEE , 1998 , 86 ( 11 ): 2278 - 2324 .

CHRABASZCZ P , LOSHCHILOV I , HUTTER F . A downsampled variant of ImageNet as an alternative to the CIFAR datasets [EB/OL ] . ( 2017-08-23 )[ 2025-09-20 ] . https://arXiv.org/abs/1707.08819 https://arXiv.org/abs/1707.08819 .

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

暂无数据