DT-EdgeGNN: Digital Twin–Driven Edge Intelligence with Graph Neural Networks for Predictive Secure VANET Communication
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
An edge intelligence-based predictive and secure communication framework Vehicular Ad Hoc Networks (VANETs) is offered based on Digital Twin (DT)-compliant Edge Intelligence (EdgeGNN). In traditional VANET there is reactive intrusion detection and centralized processing that is not able to deal with high mobility, dynamic topology and latency. The proposed solution to such obstacles ensures that the digital twin layer of real-time traffic simulation and attack case, the spatio-temporal Graph Neural Network (GNN) of vehicle interaction, and the edge intelligence layer at Right Side Unit (RSUs) are integrated to make decisions on a low-latency basis. A small blockchain system that has dynamic trust scores is embedded to guarantee safe and trustworthy communication. The framework anticipates link failure, congestion, and malicious behavior of nodes, which are created to be able to facilitate proactive routing and mitigate attacks. The results of the experiments show that the accuracy is 93.4, the attack detecting rate (0.72) is better, false positive rate (0.13) is lower and the network is much better in terms of latency and throughput. The suggested DT-EdgeGNN model can efficiently transform VANET communication into proactive management and predictive instead of reactive monitoring, which is appropriate in the next-generation advanced transportation systems.
Fields of Study & Subjects
Vehicular Ad Hoc Networks (VANETs)
Network Security and Intrusion Detection
IoT and Edge/Fog Computing
Publication details
- Journal / source
- International Journal of Computer Science and Mobile Computing
- Volume
- 15
- Issue
- 3
- Pages
- 81–90
- Publication date
- 27 March 2026
- Language
- EN