Security Prediction and Forecasting for Trust Management Systems in VANETs
Loading...
Authors
Abdelmaguid, Mohammed Anas Haroun
Date
2024-10-03
Type
thesis
Language
eng
Keyword
Network Security , VANETs , Trust Management Systems (TMSs) , Machine Learning (ML) , Proactive Security
Alternative Title
Abstract
This thesis addresses the enhancement of proactive security measures in Vehicular Ad-hoc Networks (VANETs), which are crucial for enhancing road safety and neutralizing cyber threats that can compromise human safety and network efficacy. Despite VANETs' potential to improve road safety, the predominance of reactive security measures is insufficient to combat the evolving landscape of cyber threats, and current datasets in VANETs' research are inadequate for evaluating proactive security approaches, thus stifling advancement in this crucial area. This research highlights the necessity of shifting from traditional reactive security strategies to proactive measures that anticipate and mitigate unforeseen attacks, a significant challenge with current methodologies that generally focus on known threats. Trust Management Systems (TMS) have proven effective against post-authenticated nodes. Trust is the level of confidence in accepting and acting on the received information. Trust in TMS is comprised of three components: Trust Subject, Trust Service, and Trust Origin. To enhance the detection of misbehavior messages in VANETs, we incorporate situational awareness (SA) to forecast the trustworthiness of Trust Subjects. To predict the potential impact of attacks on Trust Services, we make several contributions. We divide the attack life cycle of VANETs into proactive and retroactive phases. The proactive phase addresses the attack endgame, which refers to the impact of the attack on the physical world, such as causing accidents or creating hazards. We design an approach to predict and mitigate the attack endgame in VANETs. Furthermore, we leverage attack rate data from Trust Origin and introduce a new approach using honeypots to evaluate the network's security status by examining its ability to identify known attacks, referred to as prepared-for attacks. This information lays the groundwork to predict future attacks, which we refer to as unprepared-for attacks. Given that current datasets in VANET research are inadequate for evaluating proactive security approaches, thus impeding advancement in this crucial area, we present a new dataset, VeReMi for Attack Prediction (VeReMiAP), to evaluate the impact of attacks. This dataset aims to address the gaps in existing proactive security research tools and foster further development in proactive security strategies within VANETs. Implementation and evaluation of these techniques, utilizing open shared datasets and frameworks like VeReMi and F2MD, demonstrate the effectiveness in proactively predicting and mitigating the effects of attacks.
Description
Citation
Publisher
License
Queen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
Attribution-NonCommercial-NoDerivatives 4.0 International
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
Attribution-NonCommercial-NoDerivatives 4.0 International