Quantum-Assisted XAI-Driven DL Framework for FDI Detection in Connected Autonomous Electric Vehicles Underlying 6G
Dev Mehta, Vrutik Soni, Param Desai, Dharma Trivedi, Lakshin Pathak, Rajesh Gupta, Sudeep Tanwar · 2025
As autonomous vehicles (AVs) continue to evolve, ensuring their security against cyber threats such as False Data Injection (FDI) attacks is critical. FDI attacks manipulate sensor data, leading to unsafe driving decisions and system failures. To address this, we propose a robust detection framework using a 1D Convolutional Neural Network (1DCNN) that classifies whether incoming data is under attack. Our model enhances interpretability by integrating Explainable AI (XAI) techniques, specifically Local Interpretable Model-Agnostic Explanations (LIME), to provide insights into its decision-making process. Additionally, we leverage Vehicle-to-Everything (V2X) communication and Cooperative Perception (CP) for enhanced situational awareness while employing advanced encryption methods, including quantum encryption protocols, to ensure secure data exchange over high-speed 6G networks. Our approach significantly improves the reliability, security, and transparency of AVs, strengthening their resilience against cyber threats and fostering trust in their deployment.