A secured IoT-based intelligent transportation system using HyperGraph Neural Networks (HyperGNN) for Suspicious Activity Detection

The rapid advancement of the Internet of Things (IoT) has revolutionized Intelligent Transportation Systems (ITS), enabling real-time traffic monitoring, predictive analytics, and enhanced security. However, the increasing connectivity and data exchange in ITS pose significant security risks, including unauthorized access and suspicious activities. This paper proposes a secured IoT-based Intelligent Transportation System (ITS) utilizing HyperGraph Neural Networks (HyperGNN) for suspicious activity detection. HyperGNN is leveraged to model complex, multi-relationship data within transportation networks, capturing intricate interactions among vehicles, infrastructure, and external entities. By employing spatial and spectral hypergraph learning, the system effectively detects anomalies and malicious activities, such as unauthorized vehicle movement, cyber intrusions, and traffic violations. This security mechanism is integrated into the IoT framework to enhance real-time threat detection and mitigate potential cyber threats. Extensive simulations and real-world datasets validate the proposed approach, demonstrating superior detection accuracy, robustness, and efficiency compared to conventional GNN-based methods. The proposed HyperGNN-driven ITS enhances security, optimizes traffic management, and ensures a resilient and intelligent urban mobility system. The proposed HyperGNN achieves, 0.89 of MSE, 0.46 of RMSE and 0.39 of MAE