PG: Byzantine Fault-Tolerant and Privacy-Preserving Sensor Fusion with Guaranteed Output Delivery
Chenglu Jin, Chao Yin, Marten van Dijk, Sisi Duan, Fabio Massacci, Michael K. Reiter, Haibin Zhang
Abstract
We design and implement PG, a Byzantine fault-tolerant and privacypreserving multi-sensor fusion system. PG is flexible and extensible, supporting a variety of fusion algorithms and application scenarios. On the theoretical side, PG develops and unifies techniques from dependable distributed systems and modern cryptography. PG can provably protect the privacy of individual sensor inputs and fusion results. In contrast to prior works, PG can provably defend against pollution attacks and guarantee output delivery, even in the presence of malicious sensors that may lie about their inputs, contribute ill-formed inputs, and provide no inputs at all to sway the final result, and in the presence of malicious servers serving as aggregators. On the practical side, we implement PG in the client-serversensor setting. Moreover, we deploy PG in a cloud-based system
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Cited by top-tier papers1
Ask how each one uses itBuilds on10
- SecureML: A System for Scalable Privacy-Preserving Machine LearningPayman Mohassel, Yupeng ZhangS&P 2017 · 2,107 citations
- Global-Scale Secure Multiparty ComputationXiao Wang, Samuel Ranellucci, Jonathan KatzCCS 2017 · 220 citations
- Efficient Private Statistics with Succinct SketchesLuca Melis, George Danezis, Emiliano De CristofaroNDSS 2016 · 128 citations
- Securing Secure Aggregation: Mitigating Multi-Round Privacy Leakage in Federated LearningJinhyun So, Ramy E. Ali, Basak Güler, Jiantao Jiao et al.AAAI 2023 · 107 citations
- The Fundamental Price of Secure Aggregation in Differentially Private Federated LearningWei-Ning Chen, Christopher A. Choquette-Choo, Peter Kairouz, Ananda Theertha SureshICML 2022 · 82 citations
Related papers
- Byzantine-tolerant federated Gaussian process regression for streaming dataXu Zhang, Zhenyuan Yuan, Minghui ZhuNeurIPS 2022 · 7 citations
- Practical Differentially Private and Byzantine-resilient Federated LearningZihang Xiang, Tianhao Wang, Wanyu Lin, Di WangSIGMOD 2023 · 22 citations
- AegisFL: Efficient and Flexible Privacy-Preserving Byzantine-Robust Cross-silo Federated LearningDong Chen, Hongyuan Qu, Guangwu XuICML 2024 · 8 citations
- zPROBE: Zero Peek Robustness Checks for Federated LearningZahra Ghodsi, Mojan Javaheripi, Nojan Sheybani, Xinqiao Zhang et al.ICCV 2023 · 27 citations
- RoboRebound: Multi-Robot System Defense with Bounded-Time InteractionNeeraj Gandhi, Yifan Cai, Andreas Haeberlen, Linh Thi Xuan PhanEuroSys 2025 · 1 citation
