Post-quantum secure fog-edge computing using federated learning with blockchain

Post-quantum secure fog-edge computing using federated learning with blockchain

연구 분야: Cryptography

학회: The Journal of Supercomputing

In an era where the quantum technology revolution looms, safeguarding data security and privacy in decentralized and resource-constrained Fog-Edge environments has become the primary concern. Existing federated learning (FL) approaches, although promising for privacy-preserving model training, remain vulnerable to quantum-era attacks and face challenges in secure aggregation, trust, and data integrity. This paper addresses these challenges by proposing a novel integration of Post-Quantum Security (PQS) techniques with FL, supported by blockchain technology to enhance the resilience and security of Fog-Edge computing systems. The proposed approach leverages multivariate polynomial-based cryptography for post-quantum resilience and utilizes blockchain for immutable logging, secure aggregation, and access control of local model updates. Through extensive experimentation on real-world datasets, the method demonstrates improved performance over baseline FL systems in terms of communication efficiency, model accuracy, and quantum-resilient security guarantees. The results confirm the feasibility and effectiveness of the approach in enabling secure, privacy-preserving, and scalable FL in Fog-Edge environments under quantum threat models.