A Quantum-Optimized Graph Transformer Framework for Secure and Adaptive Trust Management in Edge Computing

This work proposes TrustFi-SecNet, a quantum-optimized graph-transformer–driven trust and security reinforcement framework designed for distributed edge–cloud environments. The model integrates multidimensional behavioral features, attention-driven trust aggregation, anomaly detection through transformer autoencoders, and chaotic lightweight encryption to ensure end-to-end security. Extensive evaluations demonstrate that TrustFi-SecNet achieves 99.1% anomaly detection accuracy, reduces average energy consumption by 66.3%, improves trust stability by 37.5%, increases secure throughput by 41.8%, and decreases latency and packet loss by 32% and 27%, respectively, compared with state-of-the-art baseline models. These results collectively establish TrustFi-SecNet as an efficient, secure, and scalable solution for modern edge intelligence ecosystems.