A Survey on Explainable Artificial Intelligence for Vehicle-to-Everything Communication Networks
Explainable Artificial Intelligence (XAI) for Vehicleto-Everything (V2X) Communication has recently attracted significant attention due to the increasing adoption of AI-driven approaches in next-generation intelligent transportation systems. AI-based techniques have been actively applied to various V2X tasks, including resource allocation, anomaly detection, and QoS provisioning. However, models such as deep reinforcement learning, federated learning, and deep neural networks often operate as black boxes, which limits their reliability and deployability in safety-critical V2X environments. To address this issue, increasing research efforts have focused on integrating XAI techniques into V2X systems in order to provide human-understandable interpretations of model decision-making processes. In this paper, we survey recent studies that apply XAI to V2X networks. In particular, we investigate the adoption of XAI techniques such as SHAP and LIME, along with their application methodologies in different V2X scenarios.