Secure and scalable blockchain-enabled 5G vehicular networks using federated trust-aware sharding and cross-shard authentication
Abstract
The rapid proliferation of fifth-generation (5G) vehicular networks requires massive connectivity and ultra-low-latency for intelligent transportation systems. While blockchain technology provides a promising decentralized foundation for data integrity and access control, existing vehicular blockchain frameworks remain limited in terms of scalability and lack integrated security mechanisms. In particular, heuristic sharding approaches such as Determinantal Point Process (DPP)-based partitioning improve throughput but ignore the trustworthiness of participating nodes and the security of cross-shard transactions. This paper proposes a Federated Trust-Aware Determinantal Point Process (FT-DPP) framework combined with a Cross-Shard Secure Transaction Channel (CS-STC) to achieve scalable and security-preserving consensus in 5G vehicular blockchains. The proposed FT-DPP integrates trust, communication quality, and transaction load into the sharding kernel, while a Proof-of-Trust (PoT) consensus assigns voting power according to node reputation. The CS-STC validates inter-shard transactions through zero-knowledge proofs (ZKP) bound to Merkle roots. Simulation and theoretical analysis indicate improved scalability, latency, and resistance to collusion compared with existing DPP-based and clustering schemes. Comprehensive simulations demonstrated that the proposed model improves shard load balance by up to 28% and reduces overall consensus delay by 20–25% compared with DPP-based and K-means benchmarks. Moreover, FT-DPP effectively lowers the attack success probability to below 5%, validating its robustness against Sybil and cross-shard spoofing attacks. The results indicated that integrating trust management with scalable sharding can provide a unified foundation for secure, efficient, and trustworthy 5G-enabled vehicular blockchains. The novelty of this work lies in introducing a trust-modulated DPP kernel that fundamentally alters the probabilistic selection behavior of shard formation. Unlike prior approaches that treat trust and diversity independently, the proposed FT-DPP integrates trust into the kernel structure itself, resulting in security-aware probabilistic selection with provable reduction in adversarial clustering.