The rapid growth of intelligent transportation systems and connected vehicle technologies has significantly increased the deployment of Internet of Vehicular Fog Computing (IoVFC) networks for real time traffic management, autonomous driving, smart parking, and accident prevention applications. Recent studies indicate that more than 75% of modern vehicles are expected to be connected to edge or fog infrastructures by 2030, while cyberattacks targeting vehicular communication networks have increased by nearly 40% over the last few years, creating serious concerns regarding data confidentiality, authentication, and network reliability. To address these limitations, this work proposes an intelligent Intrusion Detection System (IDS) for IoV driven fog computing networks using an integrated deep learning and explainable artificial intelligence (XAI) framework. Initially, the IoV dataset is collected and subjected to preprocessing techniques including normalization, missing value elimination, and feature balancing to improve data quality. Subsequently, a Quantum Graph Feature Quality Network (QG-FQN) is employed to extract discriminative spatial and relational vehicular communication features with optimized feature relevance. Finally, the extracted features are classified using the proposed XAI-Attack Detection Omni Recurrent Network (XAI-ADORN) classifier, which combines recurrent learning and XAI mechanisms to accurately identify malicious activities, improve attack interpretability, minimize false detections, and enhance secure vehicular fog communication performance.

Attack Detection, Explainable Artificial Intelligence, Internet of Vehicular Fog Computing, Intrusion Detection System, Quantum Graph Feature Quality Network, Recurrent Neural Networks