A Comprehensive Review of Intrusion Detection in In-Vehicle Networks: Advances, Challenges, and Insights from the CICIoV2024 Dataset
Rakesh Parmar, Nirav V. Bhatt · 2025
Modern connected cars raise cybersecurity concerns in In-Vehicle Networks (IVNs) and required efficient Intrusion Detection Systems (IDS). This study examines recently published CICIoV2024 and can-train-and-test datasets and IDS techniques on them with an emphasis on feature selection, machine learning (ML), deep learning (DL), and metaheuristic optimization. For attack detection optimization methods such as Principal Component Analysis (PCA), Genetic Algorithms (GA), and Particle Swarm Optimization (PSO) are analysed. Emerging technologies like Federated Learning and Blockchain-based IDS are examined, along with issues like explainability, adversarial robustness, and real-time deployment. Results indicate that hybrid machine learning techniques improve detection precision and open the door for scalable intrusion detection systems IVNs.