PQ-V2X: A Novel Post-Quantum Cryptographic Dataset for Secure Vehicular Communications

Quantum computing poses a critical threat to the cryptographic foundations of Vehicle-to-Everything (V2X) security. To address this challenge, we introduce PQ-V2X, a large-scale dataset and benchmarking suite designed for post-quantum intrusion detection. PQ-V2X integrates cryptographic artifacts, vehicular mobility traces, and zone-aware channel conditions to enable realistic evaluation of quantum-resilient intrusion detection systems (IDS). The dataset contains five million labeled samples covering 23 attack types across post-quantum, hybrid, and classical threat families, including 10.05% real-world traces derived from Kyber, Dilithium, Falcon, and SPHINCS+ implementations. We assess ten machine learning models under both synthetic and hybrid configurations and introduce the Detection Efficiency Resource (DER) metric to quantify the trade-off between detection performance and computational cost. Results show that LightGBM achieves perfect detection with superior DER while training significantly faster than CatBoost. Non-boosted models progressively narrow the classical\u2013quantum detection gap as data scales from 100K to 200K samples, although SPHINCS+-based attacks remain difficult to classify. Additionally, tunnel zones intensify detection challenges, emphasizing the importance of zone-aware features. To our knowledge, PQ-V2X is the first V2X dataset to integrate NIST post-quantum cryptographic traces, hybrid threats, realistic mobility, and resource-aware benchmarking, providing an open foundation for reproducible and scalable research on quantum-resilient V2X intrusion detection.