区块链

一种基于区块链的个性化联邦边缘学习保护框架

  • 云健 ,
  • 刘涛 ,
  • 高新茹 ,
  • 张海玉
展开
  • 大连民族大学 计算机科学与工程学院, 辽宁 大连 116650

收稿日期: 2026-01-20

  网络出版日期: 2026-10-09

Blockchain-Based Protection Framework for Personalized Federated Edge Learning

  • YUN Jian ,
  • LIU Tao ,
  • GAO Xinru ,
  • ZHANG Haiyu
Expand
  • Computer Science and Engineering College, Dalian Minzu University, Dalian 116650, Liaoning, China

Received date: 2026-01-20

  Online published: 2026-10-09

摘要

针对联邦学习在边缘场景下面临的资源约束、数据异构性及安全威胁等挑战,提出一种基于区块链的个性化联邦边缘学习保护框架。旨在通过去中心化共识机制、安全压缩与个性化微调协同作用,提升对恶意客户端及Non-IID数据的鲁棒性。通过构建安全压缩与个性化微调协同机制,实现本地模型适应能力增强、通信开销降低及聚合过程鲁棒性保障。在此基础上,设计了一种融合贡献与信任的双重激励机制,并基于Raft协议整合可验证随机函数,构建VRF-VGRaft共识算法,实现候选节点筛选、随机可验证的领导者选举及多方验证组构建。在FEMNIST和CIFAR10数据集上的仿真实验表明,相较于现有方法,该框架在处理Non-IID数据及恶意节点时,能显著提升联邦学习的安全性与效率。

本文引用格式

云健 , 刘涛 , 高新茹 , 张海玉 . 一种基于区块链的个性化联邦边缘学习保护框架[J]. 应用科学学报, 2026 , 44(5) : 731 -751 . DOI: 10.3969/j.issn.0255-8297.2026.05.003

Abstract

Federated learning in edge scenarios faces resource constraints, data heterogeneity, and security threats. This paper proposes a blockchain-based protection framework for personalized federated edge learning. The framework combines a decentralized consensus mechanism, secure compression, and personalized fine-tuning. These components jointly improve robustness to malicious clients and non-IID data. A coordination mechanism integrating secure compression and personalized fine-tuning is designed to enhance local model adaptability, reduce communication overhead, and ensure robust aggregation. A dual incentive mechanism combining contribution and trust is further proposed. Based on the Raft protocol, a verifiable random function is integrated to construct a consensus algorithm called VRF-VGRaft, enabling candidate node screening, random and verifiable leader election, and multi-party verification group construction. Simulation experiments were conducted on the FEMNIST and CIFAR10 datasets. The results show that, compared with existing methods, the framework significantly improves the security and efficiency of federated learning when handling non-IID data and malicious nodes.

参考文献

[1] Al-Sarawi S, Anbar M, Abdullah R, et al. Internet of things market analysis forecasts, 2020–2030[C]//2020 Fourth World Conference on Smart Trends in Systems, Security and Sustainability, 2020: 449-453.
[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] Pokhrel S R, Choi J. Federated learning with blockchain for autonomous vehicles: analysis and design challenges [J]. IEEE Transactions on Communications, 2020, 68(8): 4734-4746.
[4] Jiang J C, Kantarci B, Oktug S, et al. Federated learning in smart city sensing: challenges and opportunities [J]. Sensors, 2020, 20(21): 6230.
[5] Yao W, Zhou T, Han Y, et al. Verifiable secure aggregation scheme for privacy protection in federated learning networks [J]. Discover Computing, 2025, 28(1): 175.
[6] Sun S, Si C, Wu G, et al. Federated zero-shot learning with mid-level semantic knowledge transfer [J]. Pattern Recognition, 2024, 156: 110824.
[7] Feng Y, Guo Y, Hou Y, et al. A survey of security threats in federated learning [J]. Complex & Intelligent Systems, 2025, 11(2): 165.
[8] Orabi M M, Emam O, Fahmy H. Adapting security and decentralized knowledge enhancement in federated learning using blockchain technology: literature review [J]. Journal of Big Data, 2025, 12(1): 55.
[9] Kim H, Park J, Bennis M, et al. Blockchained on-device federated learning [J]. IEEE Communications Letters, 2019, 24(6): 1279-1283.
[10] Zhou S, Li K, Chen Y, et al. TrustBCFL: mitigating data bias in IoT through blockchain-enabled federated learning [J]. IEEE Internet of Things Journal, 2024, 11(15): 25648-25662.
[11] Jin R, Hu J, Min G, et al. Lightweight blockchain-empowered secure and efficient federated edge learning [J]. IEEE Transactions on Computers, 2023, 72(11): 3314-3325.
[12] Wang X, Xie Y, Chen X, et al. Securing federated learning with blockchain in the medical field: systematic literature review [J]. Journal of Medical Internet Research, 2026, 28: e79052.
[13] Lian Z, Wang W, Han Z, et al. Blockchain-based personalized federated learning for internet of medical things [J]. IEEE Transactions on Sustainable Computing, 2023, 8(4): 694-702.
[14] Huang C, Liu E, Wang R, et al. Personalized federated learning via directed acyclic graph based blockchain [J]. IET Blockchain, 2024, 4(1): 73-82.
[15] Yun J, Lu Y, Liu X. BCAFL: a blockchain-based framework for asynchronous federated learning protection [J]. Electronics, 2023, 12(20): 4214.
[16] Lin Y, Han S, Mao H, et al. Deep gradient compression: reducing the communication bandwidth for distributed training [PP/OL]. V3. (2020-06-23) [2026-01-20]. https://doi.org/10.48550/arXiv.1712.01887.
[17] Alistarh D, Grubic D, Li J, et al. QSGD: communication-efficient SGD via gradient quantization and encoding [J]. Advances in Neural Information Processing Systems, 2017:30.
[18] Bernstein J, Wang Y X, Azizzadenesheli K, et al. SignSGD: compressed optimisation for nonconvex problems [C]//International Conference on Machine Learning, 2018: 560-569.
[19] Kairouz P, Mcmahan H B. Advances and open problems in federated learning [J]. Foundations and Trends in Machine Learning, 2021, 14(1/2): 1-210.
[20] Vogels T, Karimireddy S P, Jaggi M. PowerSGD: practical low-rank gradient compression for distributed optimization [J]. Advances in Neural Information Processing Systems, 2019:32.
[21] Mugunthan V, Rahman R, Kagal L. Blockflow: an accountable and privacypreserving solution for federated learning [PP/OL]. V1. (2020-07-08) [2026-01-20]. https://doi.org/10.48550/arXiv.2007.03856.
[22] Xu M, Zou Z, Cheng Y, et al. SPDL: a blockchain-enabled secure and privacy-preserving decentralized learning system [J]. IEEE Transactions on Computers, 2022, 72(2): 548-558.
[23] Chen X, Feng C, Wang S. AIDFL: an information-driven anomaly detector for data poisoning in decentralized federated learning [J]. IEEE Access, 2025, 13: 50017-50031.
[24] Uddin M P, Xiang Y, Lu X, et al. Mutual information driven federated learning [J]. IEEE Transactions on Parallel and Distributed Systems, 2020, 32(7): 1526-1538.
[25] Xiao P, Cheng S, Stankovic V, et al. Averaging is probably not the optimum way of aggregating parameters in federated learning [J]. Entropy, 2020, 22(3): 314.
[26] 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.
[27] Chen F, Long G, Wu Z, et al. Personalized federated learning with a graph [C]//Thirty-First International Joint Conference on Artificial Intelligence, 2022: 2575-2582.
[28] Nikoloutsopoulos S, Koutsopoulos I, Titsias M K. Personalized federated learning with exact stochastic gradient descent [J]. Applied Intelligence, 2025, 55(17): 1123.
[29] Alharby M, Alssaiari A, Alateef S, et al. A quantitative analysis of the security of PoW-based blockchains [J]. Cluster Computing, 2024, 27(10): 14113-14130.
[30] Xu J, Wang C, Jia X. A survey of blockchain consensus protocols [J]. ACM Computing Surveys, 2023, 55(Suppl.13): 1-35.
[31] Abdorrahimi B, Nekouie A, Rahmani A M, et al. Blockchain technology and raft consensus for secure physician prescriptions and improved diagnoses in electronic healthcare systems [J]. Scientific Reports, 2024, 14(1): 15692.
[32] Zhou S, Ying B. Vg-Raft: an improved Byzantine fault tolerant algorithm based on raft algorithm [C]//2021 IEEE 21st International Conference on Communication Technology, 2021: 882-886.
[33] Gilad Y, Hemo R, Micali S, et al. Algorand: scaling Byzantine agreements for cryptocurrencies [C]//26th Symposium on Operating Systems Principles, 2017: 51-68.
[34] Benet J. IPFS-content addressed, versioned, P2P file system [PP/OL]. V1(2014-07-14) [2026- 01-20]. https://doi.org/10.48550/arXiv.1407.3561.
[35] Caldas S, Duddu S M K, Wu P, et al. Leaf: a benchmark for federated settings [PP/OL]. V3(2019-12-09) [2026-01-20]. https://doi.org/10.48550/arXiv.1812.01097.
[36] Li A, Sun J, Wang B, et al. LotteryFL: personalized and communication-efficient federated learning with lottery ticket hypothesis on non-IID datasets [PP/OL]. V3. (2020-08-09) [2026- 01-20]. https://doi.org/10.48550/arXiv.2008.03371.
[37] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition [C]//IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
文章导航

/

〈 〉