Efficient and Privacy-Preserving Authentication for Federated Learning in Industrial Internet of Things Data Sharing Application
- 11 December 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (9) , 11652-11663
Abstract
Federated learning (FL) has emerged as a promising approach for collaborative machine learning while respecting data privacy. However, directly aggregating local models during FL training can expose sensitive information, as adversaries may analyze model parameters to infer the underlying raw data. To address this concern, implementing a robust identity authentication mechanism can verify the integrity of local models and confirm their sources. However, many existing authentication schemes are computationally intensive, which can hinder performance. In this article, we present a lightweight authentication scheme for FL called EPPAFL. This innovative approach is compatible with all machine learning algorithms, maintains data utility, and ensures strong privacy protection. Additionally, it can be integrated with detection mechanisms to defend against data poisoning and free-rider attacks. Specifically, we first design a certificateless signature scheme based on the elliptic curve cryptography (ECC) and then propose an anonymous and efficient authentication protocol based on the Chinese remainder theorem (CRT), which supports batch verification, dynamic updates, and revocations of group keys. Moreover, participants interact anonymously to achieve secure communication, and trusted authority (TA) can track and recover the real identities of participants in case of malicious activities. We provide formal security proofs and analyses to demonstrate the achieved security properties. Performance evaluations demonstrate that our proposed protocol significantly reduces the costs associated with signature generation and verification, outperforming the existing schemes in both communication and computational efficiency.Keywords
Funding Information
- National Key Research and Development Program of China (2023YFB2703802)
- National Natural Science Foundation of China (62272120)
- Natural Science Foundation of Guangdong Province (2024A1515011426)
- Science and Technology Innovation Key Research and Development Program of Chongqing (CSTB2023TIAD-STX0031)
- Major Key Project of PCL (PCL2024A05)
This publication has 44 references indexed in Scilit:
- Lightweight Asynchronous Verifiable Secret Sharing with Optimal ResilienceJournal of Cryptology, 2024
- A quantum-secure certificateless aggregate signature protocol for vehicular ad hoc networksVehicular Communications, 2024
- PVFL: Verifiable federated learning and prediction with privacy-preservingComputers & Security, 2024
- VPFL: A verifiable privacy-preserving federated learning scheme for edge computing systemsDigital Communications and Networks, 2023
- Privacy-Preserving and Traceable Federated Learning for data sharing in industrial IoT applicationsExpert Systems with Applications, 2022
- Privacy-Preserving Federated Learning for Internet of Medical Things Under Edge ComputingIEEE Journal of Biomedical and Health Informatics, 2022
- A Blockchain-Based Multidomain Authentication Scheme for Conditional Privacy Preserving in Vehicular Ad-Hoc NetworkIEEE Internet of Things Journal, 2021
- Secure and Privacy-Preserving Federated Learning via Co-UtilityIEEE Internet of Things Journal, 2021
- Authentication and Key Management in Distributed IoT Using Blockchain TechnologyIEEE Internet of Things Journal, 2021
- Differential Privacy2006