Hybrid Quantum-Classical Computing for Intrusion Detection in Intelligent Transportation Systems
IN-PERSON – CUT202
Friday, February 16, 2024
12 PM – 1 PM ET
Abstract
Ensuring the security of information transmission in advanced traffic management systems is crucial for maintaining the integrity and reliability of traffic operations. In this research, we present a novel intrusion detection system that leverages a hybrid quantum-classical approach to enhance network security within advanced traffic management systems. By combining the strengths of quantum computing and classical techniques, our system effectively detects and mitigates intrusions in real time. We address the challenges posed by noisy quantum environments and computational overhead, developing a model that optimizes accuracy while minimizing resource demands. To comprehensively assess the capabilities of our system, we conducted rigorous evaluations using two distinct datasets: KDD99 and CICIDS. This dual-dataset approach enables a thorough evaluation of our model’s performance against both new and old attack types. Our intrusion detection system exhibits outstanding performance on the KDD99 dataset, surpassing an accuracy rate of 98.96% and an impressive 99.40% accuracy on CICIDS. In addition, our system demonstrates superior memory usage efficiency, outperforming all existing solutions in this domain. This achievement underscores our approach’s ability to maintain high accuracy while minimizing computational resource demands. These findings highlight the effectiveness of our approach in fortifying the security of advanced traffic management systems and demonstrate its potential for real-world deployment.
Presenter
Yunpeng (Jack) Zhang, PhD, is currently working as an associate professor at the University of Houston. He is the director of USDOT UTC Transportation Cybersecurity Center for Advanced Research and Education (CYBER-CARE), and the director of the Network, Software and System (NSS) Lab. Dr. Zhang’s research focuses on developing novel security and intelligence techniques to ensure cyber/physical system reliability, security, and performance in multiple industries, including transportation, energy, healthcare, commerce, finance, government, defense, the Internet of Things, etc. He is familiar with state-of-the-art research and technologies related to cyber and physical security, and artificial Intelligence, such as cryptography, access control, intrusion detection, blockchain, trust management, intelligent monitoring, deep learning, and data analysis. Dr. Zhang conducted more than 40 research and education projects supported by NSF, DoT, NIH, TxDoT, and private institutes, etc. He has worked for Boise State University (U.S.), Dakota State University (U.S.), Oak Ridge National Lab (U.S.), Imperial College London (U.K.), Queen’s University Belfast (U.K.), University of Melbourne (Australia), Northwestern Polytechnical University (China), etc. Dr. Zhang’s work has not only resulted in more than 100 publications for prestigious journals and conferences, 6 books, and 3 book chapters in the cybersecurity and software engineering fields but has also led to practical solutions to real-world problems, e.g., invented more than 70 high-performance/security new algorithms/methods, developed 30 software systems, and holds 6 patents. He supervised more than 50 graduate students. Dr. Zhang served/is serving chairs for more than 20 international conferences. He received multiple awards, including, the National Higher Education Achievement Award, which is awarded by China National Ministry of Education (The highest Education award, once every four (4) years) in Sept 2009. Dr. Zhang is keen on cooperation with academics and industry.
Previous seminars
