面向6G智能反射面辅助的联邦学习安全速率优化

武影影; 毛伯敏

doi:10.11959/j.issn.2096-109x.2025021

您当前的位置：

首页 >

文章列表页 >

面向6G智能反射面辅助的联邦学习安全速率优化

学术论文 | 更新时间：2025-05-14

- 面向6G智能反射面辅助的联邦学习安全速率优化
- Federated learning secrecy rate optimization for 6G intelligent reflecting surface assistance
- 网络与信息安全学报 2025年11卷第2期页码：115-124
- 作者机构：
  
  1.西北工业大学网络空间安全学院，陕西西安 710129
  2.西北工业大学太仓长三角研究院，江苏太仓 215400
- 作者简介：
  
  [ "武影影（1999- ），女，四川泸州人，西北工业大学博士生，主要研究方向为智能反射面、通信系统安全。" ]
  [ "毛伯敏（1989- ），男，湖北黄冈人，博士，西北工业大学教授、博士生导师，主要研究方向为通信系统安全、空天地一体化网络。" ]
- 基金信息：
  
  国家重点研发计划(2024YFF1401305)
- DOI：10.11959/j.issn.2096-109x.2025021
  中图分类号： TN915.08
- 收稿：2024-10-29，
  
  修回：2025-03-02，
  
  纸质出版：2025-04-25
- 稿件说明：
移动端阅览
武影影, 毛伯敏. 面向6G智能反射面辅助的联邦学习安全速率优化[J]. 网络与信息安全学报, 2025, 11(2): 115-124.

WU Y Y, MAO B M. Federated learning secrecy rate optimization for 6G intelligent reflecting surface assistance[J]. Chinese Journal of Network and Information Security, 2025, 11(2): 115-124.
武影影, 毛伯敏. 面向6G智能反射面辅助的联邦学习安全速率优化[J]. 网络与信息安全学报, 2025, 11(2): 115-124. DOI： 10.11959/j.issn.2096-109x.2025021.

WU Y Y, MAO B M. Federated learning secrecy rate optimization for 6G intelligent reflecting surface assistance[J]. Chinese Journal of Network and Information Security, 2025, 11(2): 115-124. DOI： 10.11959/j.issn.2096-109x.2025021.

摘要

基于非正交多址接入的联邦学习可以实现6G物联网设备的大规模接入、高速率传输及泛在智能。然而，无线信道的随机性导致模型参数传输时延波动，而频繁的模型更新可能引发通信-计算资源竞争；同时，本地模型上传阶段可能遭遇窃听攻击，导致用户隐私泄露。为了解决上述问题，提出了智能反射面辅助的基于非正交多址接入的联邦学习通信系统，旨在利用智能反射面调控无线传输环境，增加合法接收者的信号强度并抑制窃听者的接收信号，为联邦学习构建高可靠、高安全的通信基础。首先建立智能反射面辅助的通信模型；接着通过联合优化物联网设备发送功率及智能反射面各元件相移，最大化用户-基站最小安全速率；最后利用深度确定性策略梯度对上述优化问题求解，定义发送功率和智能反射面元件相移矩阵为动作空间、信道信息和上一阶段动作为状态空间。仿真结果表明，与人工噪声方法、块坐标上升方法、不部署智能反射面及不优化智能反射面元件相移相比，部署并优化智能反射面相移可以大幅提高联邦训练模型更新过程的安全速率，为后续联邦学习算法在该通信系统中的部署提供物理层保障。

Abstract

Non-orthogonal multiple access (NOMA) based federated learning (FL) can achieve the massive connectivity of internet of thing (IoT) devices

high rate transmission

and pervasive intelligence in 6G networks. However

the stochastic channels in NOMA can lead to fluctuations in the transmission delay of model parameters. Frequent model updates may cause competition for communication and computational resources

and the uploaded model can be intercepted

leading to privacy leakage. To address these issues

an intelligent reflecting surface (IRS)-assisted NOMA-based FL communication system was proposed. In this system

IRS was leveraged to reconfigure the wireless communication environment

enhancing the signal strength at the legitimate receivers and suppressing the received signal at eavesdroppers (Eves)

thereby providing a high-reliability and high-security communication foundation for federated learning. Firstly

the IRS-assisted wireless communication model was constructed. Subsequently

the minimum user-BS secrecy rate was maximized through the joint optimization of IoT device transmit power and IRS phase shifts. Finally

the deep deterministic policy gradient (DDPG) algorithm was employed to solve the optimization problem

with the action space defined as the transmit power and IRS phase shift matrix

and the channel state information and previous action constituting the state space. Simulation results demonstrated that deploying IRS in the system can greatly enhance the secrecy rate of the FL model update compared to artificial noise (AN)

the block coordinate ascent method (BCAM)

the scheme without IRS

and the scheme with random phase IRS

providing physical layer support for the subsequent deployment of FL algorithms in this communication system.

关键词

Keywords

references

廖建新 , 戚琦 , 王敬宇 , 等 . 6G智能业务网络：愿景、架构与关键技术 [J ] . 中国科学: 信息科学 , 2024 , 54 ( 5 ): 991 - 1024 .

LIAO J X , QI Q , WANG J Y , et al . 6G intelligent service networking: vision, architecture, and key technologies [J ] . Science in China (Information Sciences) , 2024 , 54 ( 5 ): 991 - 1024 .

DANG S P , AMIN O , SHIHADA B , et al . What should 6G be? [J ] . Nature Electronics , 2020 , 3 ( 1 ): 20 - 29 .

LE T H T , CANTOS L , PANDEY S R , et al . Federated learning with NOMA assisted by multiple intelligent reflecting surfaces: latency minimizing optimization and auction [J ] . IEEE Transactions on Vehicular Technology , 2023 , 72 ( 9 ): 11558 - 11574 .

MAO B M , LIU J J , WU Y Y , et al . Security and privacy on 6G network edge: a survey [J ] . IEEE Communications Surveys Tutorials , 2023 , 25 ( 2 ): 1 - 1 .

WANG T S , HUANG X M , SONG Y X , et al . Energy optimization for NOMA assisted federated learning with secrecy provisioning [C ] // 2021 IEEE/CIC International Conference on Communications in China (ICCC) . Piscataway : IEEE Press , 2021 : 1189 - 1194 .

QIAN L P , LI M Q , YE P , et al . Secrecy-driven energy minimization in federated-learning-assisted marine digital twin networks [J ] . IEEE Internet of Things Journal , 2024 , 11 ( 3 ): 5155 - 5168 .

ZHENG J , ZHANG H J , KANG J W , et al . Covert federated learning via intelligent reflecting surfaces [J ] . IEEE Transactions on Communications , 2023 , 71 ( 8 ): 4591 - 4604 .

席云光 , 李素月 . 空中智能反射面辅助毫米波通信系统能效优化 [J ] . 电讯技术 , 2024 , 64 ( 12 ): 1 - 11 .

XI Y G , LI S Y . Optimizing energy efficiency of aerial reconfigurable intelligent surface-assisted mmWave communication systems [J ] . Telecommunication Engineering , 2024 , 64 ( 12 ): 1 - 11 .

ZHU Y S , MAO B M , KATO N , Intelligent reflecting surface in 6G vehicular communications : a survey [J ] . IEEE Open Journal of Vehicular Technology , 2022 , 3 : 266 - 277 .

AN P G , YANG P , CAO X B , et al . Energy-efficient URLLC service provision via a near-space information network [J ] . IEEE Transactions on Wireless Communications , 2024 , 23 ( 8 ): 9839 - 9853 .

ZHU Y S , MAO B M , KATO N . A dynamic task scheduling strategy for multi-access edge computing in IRS-aided vehicular networks [J ] . IEEE Transactions on Emerging Topics in Computing , 2022 , 10 ( 4 ): 1761 - 1771 .

王平 , 杨志伟 , 李贺举 . 智能反射面赋能的联邦边缘学习及其在车联网中的应用 [J ] . 通信学报 , 2023 , 44 ( 10 ): 46 - 57 .

WANG P , YANG Z W , LI H J . Federated edge learning with reconfigurable intelligent surface and its application in Internet of vehicles [J ] . Journal of Communications , 2023 , 44 ( 10 ): 46 - 57 .

XIU Y , ZHAO J , YUEN C , et al . Secure beamforming for multiple intelligent reflecting surfaces aided mmWave systems [J ] . IEEE Communications Letters , 2021 , 25 ( 2 ): 417 - 421 .

WANG T S , LI Y , WU Y , et al . Secrecy driven federated learning via cooperative jamming: an approach of latency minimization [J ] . IEEE Transactions on Emerging Topics in Computing , 2022 , 10 ( 4 ): 1687 - 1703 .

CONSUL P , BUDHIRAJA I , GARG D . A hybrid secure resource allocation and trajectory optimization approach for mobile edge computing using federated learning based on Web 3.0 [J ] . IEEE Transactions on Consumer Electronics , 2024 , 70 ( 1 ): 1167 - 1179 .

WYNER A D . The wiretap channel [J ] . Bell System Technical Journal , 1975 , 54 ( 8 ): 1355 - 1387 .

ZHU Y S , MAO B M , KATO N , On a novel high accuracy positioning with intelligent reflecting surface and unscented kalman filter for intelligent transportation systems in B 5 G [J ] . IEEE Journal on Selected Areas in Communications , 2024, 42 ( 1 ): 68 - 77 .

ZHU Y S , MAO B M , KATO N , IRS-aided high-accuracy positioning for autonomous driving toward 6G : a tutorial [J ] . IEEE Vehicular Technology Magazine , 2024 , 19 ( 1 ): 85 - 92 .

SALEEM R , NI W , IKRAM M , et al . Deep-reinforcement-learning-driven secrecy design for intelligent-reflecting-surface-based 6G-IoT networks [J ] . IEEE Internet of Things Journal , 2023 , 10 ( 10 ): 8812 - 8824 .

WANG W , NI W L , TIAN H , et al . Intelligent omni-surface enhanced aerial secure offloading [J ] . IEEE Transactions on Vehicular Technology , 2022 , 71 ( 5 ): 5007 - 5022 .

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

DP-FedDRC：缓解维度坍塌的差分隐私个性化联邦学习

基于动态聚类和信誉评估的抗攻击联邦框架

基于风险权值聚合联邦学习的金融欺诈检测模型

FL-IoV：基于联邦学习的综合车联网入侵检测方法

FLWD:基于联邦学习的Webshell检测方法