The School of Computer Science is pleased to present…
Date: Monday, April 28, 2025
Time: 11:00 AM
Location: Essex Hall, Room 186
In vehicular ad hoc networks (VANETs), vehicles exchange sensitive information, including vehicle identity, position, speed, heading and other parameters, with nearby vehicles or infrastructure through Basic Safety Messages (BSM). However, VANET communication is vulnerable to various types of attacks, and appropriate security mechanisms must be in place to ensure secure transmission of BSM data. This thesis proposes transfer learning-based approaches to address these security challenges using BSM data. First, we introduce a novel transfer learning IDS that employs a CNN-BiLSTM architecture. The CNN layers, pre-trained on general network traffic data, and subsequent BiLSTM layers, trained on targeted V2V DoS attacks, significantly enhance detection accuracy and efficiency. We also present an IDS framework using the Gramian Angular Field (GAF) algorithm to transform temporal BSM sequences into image representations, enabling pre-trained Convolutional Neural Networks (CNNs) such as VGG16 to effectively detect anomalies and intrusions in vehicular communication data. Finally, we propose an ensemble framework combining multiple classifiers. Each classifier leverages transfer learning to adapt pre-trained models in the context of VANET security, enhancing overall detection performance. Our experiments utilize the VeReMiExt dataset, renowned for extensive BSM data under various attack conditions. Results show that our proposed transfer learning-based IDS approaches consistently outperform traditional and single deep-learning models, achieving higher detection rates.
Internal Reader: Dr. Imran Ahmad
Internal Reader: Dr. Boubakeur Boufama
External Reader: Dr. Ning Zhang, Department of Electrical and Computer Engineering
External Examiner: Dr. Issa Traore, Department of Electrical and Computer Engineering, University of Victoria
Advisor: Dr. Arunita Jaekel
Chair: Dr. Beth-Anne Schuelke-Leech, Department of Mechanical, Automotive and Materials Engineering
