Secure and Efficient Federated Learning Through Layering and Sharding Blockchain

被引:2
|
作者
Yuan, Shuo [1 ]
Cao, Bin [1 ]
Sun, Yao [2 ]
Wan, Zhiguo [3 ]
Peng, Mugen [1 ]
机构
[1] Beijing Univ Posts & Telecommun, State Key Lab Networking & Switching Technol, Beijing 100876, Peoples R China
[2] Univ Glasgow, James Watt Sch Engn, Glasgow G12 8QQ, Scotland
[3] Zhejiang Lab, Hangzhou 311121, Peoples R China
基金
国家重点研发计划;
关键词
Blockchains; Training; Internet of Things; Computational modeling; Computer architecture; Security; Throughput; Federated learning; blockchain; direct acyclic graph; sharding; layering; PRIVATE;
D O I
10.1109/TNSE.2024.3361458
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Introducing blockchain into Federated Learning (FL) to build a trusted edge computing environment for transmission and learning has attracted widespread attention as a new decentralized learning pattern. However, traditional consensus mechanisms and architectures of blockchain systems face significant challenges in handling large-scale FL tasks, especially on Internet of Things (IoT) devices, due to their substantial resource consumption, limited transaction throughput, and complex communication requirements. To address these challenges, this paper proposes ChainFL, a novel two-layer blockchain-driven FL system. It splits the IoT network into multiple shards within the subchain layer, effectively reducing the scale of information exchange, and employs a Direct Acyclic Graph (DAG)-based mainchain as the mainchain layer, enabling parallel and asynchronous cross-shard validation. Furthermore, the FL procedure is customized to integrate deeply with blockchain technology, and a modified DAG consensus mechanism is designed to mitigate distortion caused by abnormal models. To provide a proof-of-concept implementation and evaluation, multiple subchains based on Hyperledger Fabric and a self-developed DAG-based mainchain are deployed. Extensive experiments demonstrate that ChainFL significantly surpasses conventional FL systems, showing up to a 14% improvement in training efficiency and a threefold increase in robustness.
引用
收藏
页码:3120 / 3134
页数:15
相关论文
共 50 条
  • [1] DAG-Based Blockchain Sharding for Secure Federated Learning with Non-IID Data
    Lee, Jungjae
    Kim, Wooseong
    [J]. SENSORS, 2022, 22 (21)
  • [2] ScaleSFL: A Sharding Solution for Blockchain-Based Federated Learning
    Madill, Evan
    Nguyen, Ben
    Leung, Carson K.
    Rouhani, Sara
    [J]. BSCI'22: PROCEEDINGS OF THE FOURTH ACM INTERNATIONAL SYMPOSIUM ON BLOCKCHAIN AND SECURE CRITICAL INFRASTRUCTURE, 2022, : 95 - 106
  • [3] Lightweight Blockchain-Empowered Secure and Efficient Federated Edge Learning
    Jin, Rui
    Hu, Jia
    Min, Geyong
    Mills, Jed
    [J]. IEEE TRANSACTIONS ON COMPUTERS, 2023, 72 (11) : 3314 - 3325
  • [4] Secure and Efficient Blockchain-Based Federated Learning Approach for VANETs
    Asad, Muhammad
    Shaukat, Saima
    Javanmardi, Ehsan
    Nakazato, Jin
    Bao, Naren
    Tsukada, Manabu
    [J]. IEEE INTERNET OF THINGS JOURNAL, 2024, 11 (05): : 9047 - 9055
  • [5] Permissioned Blockchain Frame for Secure Federated Learning
    Sun, Jin
    Wu, Ying
    Wang, Shangping
    Fu, Yixue
    Chang, Xiao
    [J]. IEEE COMMUNICATIONS LETTERS, 2022, 26 (01) : 13 - 17
  • [6] Federated Learning Meets Blockchain to Secure the Metaverse
    Moudoud, Hajar
    Cherkaoui, Soumaya
    [J]. 2023 INTERNATIONAL WIRELESS COMMUNICATIONS AND MOBILE COMPUTING, IWCMC, 2023, : 339 - 344
  • [7] Resource-Efficient Federated Learning and DAG Blockchain With Sharding in Digital-Twin-Driven Industrial IoT
    Jiang, Li
    Liu, Yi
    Tian, Hui
    Tang, Lun
    Xie, Shengli
    [J]. IEEE INTERNET OF THINGS JOURNAL, 2024, 11 (10): : 17113 - 17127
  • [8] Secure Data Sharing in Federated Learning through Blockchain-Based Aggregation
    Liu, Bowen
    Tang, Qiang
    [J]. FUTURE INTERNET, 2024, 16 (04)
  • [9] DRL-Based Adaptive Sharding for Blockchain-Based Federated Learning
    Lin, Yijing
    Gao, Zhipeng
    Du, Hongyang
    Kang, Jiawen
    Niyato, Dusit
    Wang, Qian
    Ruan, Jingqing
    Wan, Shaohua
    [J]. IEEE TRANSACTIONS ON COMMUNICATIONS, 2023, 71 (10) : 5992 - 6004
  • [10] BASS: A Blockchain-Based Asynchronous SignSGD Architecture for Efficient and Secure Federated Learning
    Xu, Chenhao
    Ge, Jiaqi
    Deng, Yao
    Gao, Longxiang
    Zhang, Mengshi
    Li, Yong
    Zhou, Wanlei
    Zheng, Xi
    [J]. IEEE Transactions on Dependable and Secure Computing, 2024, 21 (06) : 5388 - 5402