Access Type
Open Access Embargo
Date of Award
January 2024
Degree Type
Dissertation
Degree Name
Ph.D.
Department
Electrical and Computer Engineering
First Advisor
Lubna Alazzawi
Second Advisor
Harpreet Singh
Abstract
The integration of cloud-based technologies into Connected and Autonomous Vehicles (CAVs) is reshaping the field by combining Deep Federated Learning (DFL), Security Information and Event Management (SIEM), and cloud-dew computing. This solution leverages cloud-based resource provisioning, which is crucial for allocating scalable and efficient computational resources in a dynamic manner. These resources are essential for managing the intricate data and computing requirements of distributed systems, especially in the intelligent vehicle sector. This provisioning facilitates the efficient control of route mapping and cybersecurity in Connected Autonomous Vehicles (CAVs), guaranteeing the ability to process and make decisions in real-time.The research evaluates the reliability of cloud-based frameworks in Connected and Autonomous Vehicles (CAVs), focusing on using dynamic maps for generating secure, real-time paths. This approach highlights the importance of cloud technologies in updating dynamic map data for intelligent transportation systems. Central to this study is the cloud- dew computing architecture, which merges cloud computing's extensive storage and processing capabilities with dew computing's real-time, localized data handling at the end user's edge. This two-tier system allows the cloud layer to manage large-scale processing and analytics, while the dew layer, closer to CAVs, handles immediate data processing tasks. This setup facilitates swift responses to real-time data, essential for the dynamic needs of autonomous vehicles. The seamless integration of both layers ensures efficient and timely updates to CAV navigation and cybersecurity, enhancing both data processing speed and security, a crucial aspect for the safety and reliability of autonomous vehicles and Cyber-Physical Systems (CPS). State-of-the-art deep learning techniques, including a custom Stacked Autoencoder and a Long Short-Term Memory Autoencoder (LSTM-AE), for cybersecurity and Intrusion Detection Systems (IDS). This model demonstrates exceptional proficiency in managing time-series data for self-governing systems, showcasing high accuracy and adaptability for on-road applications. Furthermore, the utilization of distributed processing is necessary for path planning, as it involves multiple cyber-physical systems. This entails employing both synchronous and asynchronous data parallelism approaches under the cloud-dew computing paradigm. An essential aspect of this study entails a comprehensive analysis of different processing units, evaluating their capacity to be scaled, their dependability, and their effectiveness. This assessment is vital for measuring the effectiveness of reducing training durations and optimizing the overall performance of the system. The selected methodology is crucial for protecting CAVs from rising cyber risks and guaranteeing the security and dependability of automated path planning, a vital element in the autonomous vehicle industry.
Recommended Citation
Moussa, Mohamed Mounir, "Security Information And Event Management Optimization Using Deep Federated Learning In Cloud-Based Autonomous Cyber-Physical Systems" (2024). Wayne State University Dissertations. 4121.
https://digitalcommons.wayne.edu/oa_dissertations/4121