文章详情

题名 · 作者 · 摘要 · 关键词 · DOI · 参考文献

联邦学习隐私保护与安全防御策略研究综述

袁刘，张永，焦鹏举，仲阳，滕圣波

（中移（苏州）软件技术有限公司，江苏 苏州 215000）

栏目：网络与信息安全 · 第 2026 年第 17 期 · 页码：120-128,136 · 2096-4706（2026）17-0120-10 · 中图分类号：TP18 · 文献标识码：A · DOI：10.19850/j.cnki.2096-4706.2026.17.022

摘要

联邦学习作为一种分布式机器学习方法，通过加密参数交换，在保持数据留存本地的前提下将分散的数据实现联合训练，同时兼顾数据共享和隐私保护，现已广泛应用于物联网、工业物联网、智能医疗、电力系统等领域。然而，联邦学习依旧存在潜在的隐私泄露隐患和数据安全问题。文章系统梳理了联邦学习存在的隐私和安全问题，分析了多种恶意攻击手段和安全防御手段，以帮助规避在构建联邦学习计算环境时可能面临的潜在风险。同时，文章还探讨了联邦学习的未来研究方向，以期为进一步推动联邦学习在复杂场景下的实践应用提供参考。

关键词

联邦学习；隐私保护；数据安全；机器学习

参考文献

  1. [1] Sajjad M，Nasir M，Muhammad K，et al. Raspberry Pi Assisted Face Recognition Framework for Enhanced Law-enforcement Services in Smart Cities [J].Future Generation Computer Systems，2020，108：995-1007.
  2. [2] McMahan B，Moore E，Ramage D，et al. Communication-efficient Learning of Deep Networks from Decentralized Data [C]//Artificial Intelligence and Statistics，2017：1273-1282.
  3. [3] Wang P，Liu B，Zeng D，et al. Personalized Federated Learning via Backbone Self-distillation [C]//Proceedings of the 5th ACM International Conference on Multimedia in Asia，2023：1-7.
  4. [4] Wei K，Li J，Ding M，et al. Federated Learning with Differential Privacy：Algorithms and Performance Analysis [J].IEEE Transactions on Information Forensics and Security，2020，15：3454-3469.
  5. [5] 黄精武.基于差分隐私的联邦学习数据隐私安全技术 [J].通信技术，2022，55（12）：1618-1625.
  6. [6] 王超，龙士工，刘光源，等.面向隐私保护联邦学习的激励机制 [J].计算机工程与设计，2024，45（12）：3521-3530.
  7. [7] Ye M，Fang X，Du B，et al. Heterogeneous Federated Learning: State-of-the-art and Research Challenges[J]. ACM Computing Surveys，2023，56（3）：1-44.
  8. [8] Deng Y，Lyu F，Ren J，et al. FAIR: Quality-aware Federated Learning with Precise User Incentive and Model Aggregation [C]//2021 IEEE INFOCOM 2021-IEEE Conference on Computer Communications，2021：1-10.
  9. [9] Gkillas A，Lalos A S. Missing Data Imputation for Multivariate Time Series in Industrial IoT: A Federated Learning Approach [C]//2022 IEEE 20th International Conference on Industrial Informatics (INDIN)，2022：87-94.
  10. [10] Novikova E S，Chen Y，Meleshko A V. Evaluation of Data Heterogeneity in FL Environment [C]//2024 XXVII International Conference on Soft Computing and Measurements (SCM)，2024：344-347.
  11. [11] Ribero M，Vikalo H，De V G. Federated Learning under Intermittent Client Availability and Time-varying Communication Constraints [J].IEEE Journal of Selected Topics in Signal Processing，2023，17 （1）：98-111.
  12. [12] Ilhan F，Su G，Liu L. Scalefl: Resource-adaptive Federated Learning with Heterogeneous Clients [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2023：24532-24541.
  13. [13] Wang Z，Xu H，Liu J，et al. Resource-efficient Federated Learning with Hierarchical Aggregation in Edge Computing [C]//IEEE INFOCOM 2021-IEEE Conference on Computer Communications，2021：1-10.
  14. [14] Jiang J C，Kantarci B，Oktug S，et al. Federated Learning in Smart City Sensing: Challenges and Opportunities [J].Sensors，2020，20（21）：6230.
  15. [15] Xu C，Qu Y，Eklund P W，et al. BAFL: An Efficient Blockchain-based Asynchronous Federated Learning Framework [C]//2021 IEEE Symposium on Computers and Communications (ISCC)，2021：1-6.
  16. [16] Jiang Z，Xu Y，Xu H，et al. Computation and Communication Efficient Federated Learning with Adaptive Model Pruning [J].IEEE Transactions on Mobile Computing，2024，23（3）：2003-2021.
  17. [17] Moore E，Imteaj A，Rezapour S，et al. A Survey on Secure and Private Federated Learning Using Blockchain: Theory and Application in Resource-constrained Computing [J].IEEE Internet of Things Journal，2023，10（24）：21942-21958.
  18. [18] Imteaj A，Thakker U，Wang S Q，et al. A Survey on Federated Learning for Resource-Constrained IoT Devices [J].IEEE Internet of Things Journal，2022，9（1）：1-24.
  19. [19] Shah S M，Lau V K N. Model Compression for Communication Efficient Federated Learning [J].IEEE Transactions on Neural Networks and Learning Systems，2023，34（9）：5937-5951.
  20. [20] Wang H N，Liu Z L，Hoshino K，et al. FedPaI: Achieving Extreme Sparsity in Federated Learning via Pruning at Initialization [PP/OL].arXiv:2504.00308 [cs.LG].https://arxiv.org/abs/2504.00308.
  21. [21] 马振国，和孜轩，孙彦景，等.基于自适应模型剪枝的联邦遗忘学习方法研究 [J].电子与信息学报，2025，47（11）：4515-4524.
  22. [22] Büyüktaş B，Klotz J，Demir B. FedX: Explanation-Guided Pruning for Communication-Efficient Federated Learning in Remote Sensing [J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2026，19：8549-8560.
  23. [23] Alistarh D，Grubic D，Li J，et al. QSGD: Communication-efficient SGD via Gradient Quantization and Encoding [C]//Advances in Neural Information Processing Systems，2017，30：1709-1720.
  24. [24] Lin X H，Liu Y，Chen F J，et al. Stochastic Gradient Compression for Federated Learning over Wireless Network [J].China Communications，2024，21（4）：230-247.
  25. [25] Kim S W，Kim S，Kim J，et al. FedWSQ: Efficient Federated Learning with Weight Standardization and Distribution-Aware Non-Uniform Quantization [C]//2025 IEEE/CVF International Conference on Computer Vision (ICCV)，2025：4616-4625.
  26. [26] He Y T，Chen Y Q，Yang X D，et al. Learning Critically: Selective Self-Distillation in Federated Learning on Non-IID Data [J].IEEE Transactions on Big Data，2024，10：789-800.
  27. [27] Song W S，Yan M W，Li X Z，et al. Bidirectional Decoupled Distillation for Heterogeneous Federated Learning [J/OL].Entropy，2024，26（9）：762.https://doi.org/10.3390/e26090762.
  28. [28] 陈学斌，任志强.PFKD：综合考虑数据异构和模型异构的个性化联邦学习框架 [J].南京信息工程大学学报，2024，16（4）：513-519.
  29. [29] Guastella A，Sani L，Iacob A，et al. SparsyFed: Sparse Adaptive Federated Training [PP/OL].arXiv:2504.05153 [cs.LG].https://arxiv.org/abs/2504.05153.
  30. [30] Mai P H，Ding Y L，Lyu Z Y，et al. SecEmb: Sparsity-Aware Secure Federated Learning of On-Device Recommender System with Large Embedding [C]//Proceedings of the 42nd International Conference on Machine Learning，2025：42642-42667.
  31. [31] Jiang C，Zhou H，Zhang X，et al. FedAR: Addressing Client Unavailability in Federated Learning with Local Update Approximation and Rectification [C]//Machine Learning and Knowledge Discovery in Databases，2024：178-196.
  32. [32] Song J J，Luo J J，Lu R W，et al. A Joint Approach to Local Updating and Gradient Compression for Efficient Asynchronous Federated Learning [C]//International Conference on Parallel Processing，2024：196-211.
  33. [33] Tang X，Peng R，Li H，et al. AchiFL: Efficient Asynchronous Hierarchical Federated Learning via Adaptive Edge Aggregation [J].IEEE Transactions on Cognitive Communications and Networking，2026，12：4746-4761.
  34. [34] Lu C Y，Sun Y D，Yang Z C，et al. FedPSA: Modeling Behavioral Staleness in Asynchronous Federated Learning [PP/OL].arXiv:2602.15337 [cs.LG].https://arxiv.org/abs/2602.15337.
  35. [35] Egger M，Bitar R，Wachter-Zeh A，et al. BICompFL: Stochastic Federated Learning with Bi-Directional Compression [PP/OL].arXiv:2502.00206 [cs.LG].https://arxiv.org/abs/2502.00206.
  36. [36] Ye Z J，Di S，Wang J M，et al. An Efficient Gradient-Aware Error-Bounded Lossy Compressor for Federated Learning [PP/OL].arXiv:2511.05770 [cs.LG].https://arxiv.org/abs/2511.05770.
  37. [37] 冯奕铭，钱珍，李光辉，等.异构边缘环境下自适应分层联邦学习协同优化方法 [J].计算机研究与发展，2025，62（6）：1416-1433.
  38. [38] Guo Z H，Liu Z Y，Cen Z H，et al. CoCo-Fed: A Unified Framework for Memory- and Communication-Efficient Federated Learning at the Wireless Edge [PP/OL].arXiv:2601.00549 [cs.IT].https://arxiv.org/abs/2601.00549.
  39. [39] Mu Y J，Shen C. Federated Split Learning with Improved Communication and Storage Efficiency [J].IEEE Transactions on Mobile Computing，2026，25（1）：272-283.
  40. [40] Dong A，Chen W N，Ozgur A. Leveraging Randomness in Model and Data Partitioning for Privacy Amplification [PP/OL].arXiv:2503.03043 [cs.LG].https://arxiv.org/abs/2503.03043.
  41. [41] Chen L，Liu W，Chen Y，et al. Communication-efficient Design for Quantized Decentralized Federated Learning [J].IEEE Transactions on Signal Processing，2024，72：1175-1188.
  42. [42] Liu F X，Zheng Z M，Shi Y X，et al. A Survey on Federated Learning: A Perspective from Multi-party Computation [J/OL].Frontiers of Computer Science，2024，18（1）：181336.https://doi.org/10.1007/s11704-023-3282-7.
  43. [43] Ammad-ud-din M，Ivannikova E，Khan S A，et al. Federated Collaborative Filtering for Privacy-Preserving Personalized Recommendation System [PP/OL].arXiv:1901.09888 [cs.IR].https://arxiv.org/abs/1901.09888.
  44. [44] Zhao X H，Zheng Y M，Wan J X，et al. FedDyH: A Multi-Policy with GA Optimization Framework for Dynamic Heterogeneous Federated Learning [J/OL].Biomimetics，2025，10（3）：185.https://doi.org/10.3390/biomimetics10030185.

作者简介

袁刘（1990.01—），男，汉族，江苏宿迁人，中级工程师，硕士研究生，研究方向：云计算与人工智能。