Privacy-aware distributed intelligence with tokenized trust for low-latency task offloading in 6G vehicular edge...

Area of Science:

• * Next-generation 6G vehicular networks
• * Distributed artificial intelligence
• * Edge computing

Background:

• * 6G vehicular networks demand ultra-low latency, secure cooperation, and efficient task offloading.
• * Existing systems often optimize latency or energy in isolation, neglecting joint privacy and long-term trust.
• * Vehicle-to-vehicle (V2V) edge computing faces challenges in privacy preservation and trust management.

Purpose of the Study:

• * To propose a distributed intelligence architecture for V2V edge computing in 6G networks.
• * To enable privacy-preserving collaborative prediction and decentralized incentive enforcement.
• * To address task allocation as a multi-objective optimization problem considering latency, energy, communication stability, and privacy.

Main Methods:

• * Integration of federated learning (FL) for privacy-preserving collaborative prediction.
• * Implementation of blockchain-based trust management for decentralized incentive enforcement and node security.
• * Application of a learning-coupled primal-dual optimization for multi-objective task allocation.

Main Results:

• * Achieved up to 30-40% reduction in service latency.
• * Demonstrated approximately 25% improvement in task completion rate.
• * Enhanced privacy preservation through gradient-based learning and achieved up to 95% accuracy in detecting malicious nodes.

Conclusions:

• * The proposed framework effectively integrates FL and blockchain for scalable, privacy-aware, and trustworthy distributed intelligence in 6G vehicular edge networks.
• * The joint optimization approach successfully balances competing objectives of latency, energy, stability, and privacy.
• * The system demonstrates superior performance compared to state-of-the-art baselines in simulated traffic-network-blockchain environments.