广东工业大学学报 ›› 2021, Vol. 38 ›› Issue (06): 47-52,83.doi: 10.12052/gdutxb.210114

• • 上一篇    下一篇

面向6G网络的联邦学习资源协作激励机制设计

蒋丽1, 谢胜利1, 张彦2   

  1. 1. 广东工业大学 自动化学院，广东 广州 510006;
    2. 挪威奥斯陆大学 信息学院，奥斯陆 0316
  • 收稿日期:2021-07-14 出版日期:2021-11-10 发布日期:2021-11-09
  • 通信作者: 张彦，男，教授，博士，欧洲科学院院士，主要研究方向为6G网络和绿色通信等
  • 作者简介:蒋丽(1986–)，女，特聘副教授，主要研究方向为6G网络和网络内生安全等，E-mail：jiangli@gdut.edu.cn
  • 基金资助:
    国家重点研发计划项目(2020YFB1807801)；移动通信教育部工程研究中心开放研究项目(cqupt-mct-202003)

Incentivizing Resource Cooperation for Federated Learning in 6G Networks

Jiang Li1, Xie Sheng-li1, Zhang Yan2   

  1. 1. School of Automation, Guangdong University of Technology, Guangzhou 510006, China;
    2. Department of Informatics, University of Oslo, Oslo 0316, Norway
  • Received:2021-07-14 Online:2021-11-10 Published:2021-11-09

HTML

PDF (PC)

2620

摘要： 第6代(6G)移动通信在满足人类智能通信需求的同时, 也给用户数据的安全与隐私保护带来了极大挑战。为此, 基于联邦学习的分布式机器学习架构应运而生。然而, 在联邦的模型训练过程中, 移动设备会产生大量计算和通信开销。自私的移动设备不愿意参与模型训练, 这将降低联邦学习性能。本文基于迭代双边拍卖设计了一种有效的联邦学习资源协作激励机制, 任务计算终端作为卖方, 任务请求终端作为买方, 本地接入点根据买卖双方的出价做出模型训练时延和相应定价决策, 在买卖双方信息非对称情况下最大化联邦学习市场总效用。仿真实验表明, 所提机制具有良好的收敛性, 可显著提高联邦学习的准确率, 同时降低训练损失。

关键词: 6G网络, 安全与隐私保护, 联邦学习, 资源协作, 迭代双边拍卖

Abstract: Recent advances in 6th Generation (6G) mobile networks can meet the needs of deeper intelligent communication. Meanwhile, it also brings great challenges to the security and privacy preservation of user data. Federated learning is emerging as a distributed learning method to preserve privacy by enabling users to train machine learning models locally and requiring the users to upload only model parameters instead of sending original data to the server. However, the users with mobile devices may be unwilling to participate in federated learning tasks due to considerable overhead for computation and communication, which degrades performance and hinders the application of federated learning. In this study, an iterative double auction is designed based on resource cooperation scheme to incentivize the mobile devices to contribute their resources in federated learning tasks, where model trainers act as sellers and task requesters as buyers. Access point determines optimal training time cost and payment according to the prices offered by the sellers and the buyers. The goal of our incentivizing resource cooperation scheme is to maximize the total utility of federated learning market under information asymmetry. Numerical results show that the proposed scheme can converge to the optimal solution, and also can significantly improve model accuracy and degrade model loss value.

Key words: 6G mobile networks, security and privacy preservation, federated learning, resource cooperation, iterative double auction

中图分类号: 

  • TN929.5
[1] 索士强, 王映民. 未来6G网络内生智能的探讨与分析[J]. 移动通信, 2020, 44(6): 126-30.
SUO S Q, WANG Y M. Discussion and analysis on the native intelligence for the future 6G network [J]. Mobile Communications, 2020, 44(6): 126-30.
[2] 孙耀华, 王则予, 袁硕, 等. 智能内生6G网络: 架构、用例和挑战[J]. 电子技术应用, 2021, 47(3): 8-13.
SUN Y H, WANG Z Y, YUAN S, et al. The sixth-generation mobile communication network with endogenous intelligence architectures, use cases and challenges [J]. Application of Electronic Technique, 2021, 47(3): 8-13.
[3] 郑思远, 崔苗, 张广驰. 基于强化学习的无人机安全通信轨迹在线优化策略[J]. 广东工业大学学报, 2021, 38(4): 60-64.
ZHENG S Y, CUI M, ZHANG G C. Reinforcement learning-based online trajectory optimization for secure UAV communications [J]. Journal of Guangdong University of Technology, 2021, 38(4): 60-64.
[4] 张成磊, 付玉龙, 李晖, 等. 6G网络安全场景分析及安全模型研究[J]. 网络与信息安全学报, 2021, 7(1): 29-45.
ZHANG C L, FU Y L, LI H, et al. Research on security scenarios and security models for 6G networking [J]. Chinese Journal of Network and Information Security, 2021, 7(1): 29-45.
[5] KONECNY J, MCMAHAN H B, RAMAGE D. Federated optimization: distributed machine learning for on-device intelligence[EB/OL]. [2016-10-08]. https://arxiv.org/abs/1610.02527v1.
[6] HARD A, RAO K, MATHEWS R, et al. Federated learning for mobile keyboard prediction [EB/OL]. [2019-02-28]. https://arxiv.org/abs/1811.03604v2.
[7] CHEN M, YANG Z, SAAD W, et al. A joint learning and communications framework for federated learning over wireless networks [J]. IEEE Transactions on Wireless Communications, 2021, 20(1): 269-283.
[8] NI W, LIU Y, TIAN H. Intelligent reflecting surfaces enhanced federated learning[C]//IEEE Globecom Workshops. Taipei: IEEE, 2021: 7-11.
[9] LU Y, HUANG X, ZHANG K, et al. Blockchain empowered asynchronous federated learning for secure data sharing in internet of vehicles [J]. IEEE Transactions on Vehicular Technology, 2020, 69(4): 4298-4311.
[10] LE T, TRAN N, TUN Y, et al. An incentive mechanism for federated learning in wireless cellular network: an auction approach [J]. IEEE Transactions on Wireless Communications, 2021, 20(8): 4874-4887.
[11] SUN W, WANG P, XU N, et al. Dynamic digital twin and distributed incentives for resource allocation in aerial-assisted internet of vehicles[J/OL]. IEEE Internet of Things Journal (Early Access) [2021-07-01]. https://ieeexplore.ieee.org/document/9351542. DOI: 10.1109/JIOT.2021.3058213.
[12] TRAN N, BAO W, ZOMAYA A, et al. Federated learning over wireless networks: optimization model design and analysis [C]//IEEE Infocom 2019-IEEE Conference on Computer Communications. Paris: IEEE, 2019: 1387-1395.
[13] FRIEDMAN D, RUST J. The double auction market: institutions theories and evidence [M]. Boulder: Westview Press, 1993: 429.
[14] BOYD S, VANDENBERGHE L. Convex optimization [M]. New York: Cambridge University Press, 2004: 727.
[15] KRIZHEVSKY A. Learning multiple layers of features from tiny images [D]. Toronto: University of Toronto, 2009.
[1] 吴庆捷, 崔苗, 张广驰, 陈伟. 无人机信息采集系统的端到端吞吐量最大化研究[J]. 广东工业大学学报, 2022, 39(06): 53-61.
[2] 王丰, 李宇龙, 林志飞, 崔苗, 张广驰. 基于计算吞吐量最大化的能量采集边缘计算系统在线资源优化配置[J]. 广东工业大学学报, 2022, 39(04): 17-23.
[3] 吴家锐, 崔苗, 张广驰, 王丰. 可重构智能表面辅助的非正交多址接入系统的安全通信研究[J]. 广东工业大学学报, 2022, 39(03): 49-54,69.
[4] 郑思远, 崔苗, 张广驰. 基于强化学习的无人机安全通信轨迹在线优化策略[J]. 广东工业大学学报, 2021, 38(04): 59-64.
[5] 谭艺枝, 陈宝仁. 无线携能传输协同中继非正交多址接入系统的速率优化设计[J]. 广东工业大学学报, 2020, 37(01): 81-86.
[6] 崔苗, 庞浩然, 张广驰, 刘怡俊, 邹为民. 基于无线供电的非正交多址接入网络的鲁棒资源分配研究[J]. 广东工业大学学报, 2018, 35(06): 37-42.
[7] 童辉志, 张广驰, 周绪龙, 崔苗, 刘怡俊, 林凡. 具有能量获取基站的相邻多蜂窝小区的能量与频谱分配研究[J]. 广东工业大学学报, 2018, 35(04): 68-74.
[8] 蒋阅, 刘海林, 王强. 基于基站休眠技术的无线通信网络能源协作策略[J]. 广东工业大学学报, 2018, 35(02): 69-74.
[9] 何丽君, 张广驰, 黄高飞, 万林青, 崔苗, 刘怡俊, 林凡. 无线携能OFDM中继系统的联合资源分配研究[J]. 广东工业大学学报, 2018, 35(01): 29-34.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发