Post-Quantum Secure Blockchain-Based Federated Learning Framework for Healthcare Analytics
Daniel Commey, Sena Hounsinou, Garth V. Crosby
Abstract
The growth of IoT in healthcare generates massive sensitive data. This necessitates a secure and privacy-preserving distributed network to transport and process the data. Federated learning (FL) offers privacy-preserving model training, while blockchain ensures data integrity through transparency and immutability. Yet, quantum computing threatens cryptographic schemes like ECDSA, endangering long-term data confidentiality. This paper integrates post-quantum cryptography (PQC) with blockchain-based FL for healthcare analytics. We evaluate three signature-based PQC algorithms–Falcon, Dilithium (ML-DSA-65), and SPHINCS+ (SPHINCS+-SHA2-128s)–to assess their impact on blockchain transaction costs and latency. Benchmarks on a local Ethereum testnet show that lattice-based schemes, particularly ML-DSA-65, achieve verification under 10 ms with acceptable gas costs. Our findings indicate that smart contract signature verification is the primary gas consumer, offering guidelines for deploying quantum-resistant FL systems. These findings justify and potentially create a foundation for building complete systems that integrate PQC into Blockchain-based FL systems.