System for Detecting Intrusions in the Internet of Vehicles using Deep Learning
Mrs. Vibha MP, Prof. Dilna PM · International Journal of Research Publication and Reviews · 2025
Modern vehicles, including both connected and autonomous types, are progressively integrating with external networks, offering a wide array of services and functionalities.But this increased connectedness also makes the Internet of Vehicles (IoV) more open to cyberattacks and more vulnerable.Systems for intrusion detection (IDSs) are essential for protecting sophisticated automobile systems from network attacks because automotive networks lack authentication and encryption.This paper introduces two IDS that leverages ensemble learn-ing and transfer learning, incorporating convolutional neural networks (CNNs) and hyper parameter optimization techniques for IoV systems.and a hybrid model consisits of CNN and LSTM.My experiments demonstrate that the proposed ensemble IDS achieves detection rates and F1-scores exceeding 99.94% on the widely recognized public benchmark IoV security dataset, CICIDS2017, and Proposed hybrid model achieves detection rates and F1-scores exceeding 99.97% on the same dataset.This highlights the effectiveness of the proposed IDS in identifying cyber attacks within external vehicular networks.