报告题目：Towards Secure Remote Monitoring of Cyber-Physical Systems:A Set-Membership State Estimation Approach
报告人：Dr. Xiaohua Ge Swinburne University of Technology, Melbourne, Australia
报告及会议研讨时间：2020年10月28日（周三）08:00-12:00
参与方式：腾讯会议ID:994384472
Abstract:
Cyber-physical systems (CPSs) are broadly employed to provide remote monitoring and control solutions for critical infrastructure, including electrical power grids, water, oil and gas supply distribution systems. Meanwhile, modern infrastructure faces constant security threats and attacks against its communication-control systems such as supervisory control and data acquisition (SCADA) systems. Undoubtedly, the operation of modern critical infrastructure relies on secure state estimation mechanisms to enable reliable remote monitoring of the geographically dispersed assets.
In this talk, we introduce a set-membership state estimation approach to achieve resilient and secure monitoring of networked systems against a variety of cyber-physical attacks. We first start with a brief background on set-membership estimation. Second, we present some preliminary yet promising security results on ellipsoidal set-membership estimation of the concerned system when attacks are confined to some unknown but bounded set. Several case studies of a SCADA water supply distribution system are provided to allow comprehensive analysis and evaluation of different attack scenarios. Third, we demonstrate that the presented ellipsoidal set-membership estimation framework can be further used to offer resilience and security guarantees against certain type of attacks even without any a priori knowledge of the attacks. Different cyber-physical attacks on a simple SCADA water supply distribution system are also examined. Finally, we highlight some challenging issues worthy of further research.
Biography
Dr. Xiaohua Ge is currently a Senior Lecturer with the School of Software and Electrical Engineering, Swinburne University of Technology, Melbourne, VIC, Australia. His research is oriented towards intelligent, resilient, secure and privacy-preserving estimation and control theories for multi-agent systems and cyber-physical systems and their applications. He has yet published fifty fully-refereed high quality journal articles including over twenty in the most prestigious IEEE Transactions and Automatica. He has been awarded as the 2019 Web of Science Highly Cited Researcher, and also received the 2019 IEEE SMC Andrew P. Sage Best Transactions Paper Award. Hehasserved or is serving as a Guest Editor of a number of prestigious journals, including the IEEE TRANSACTION ON SYSTEMS, MAN, AND CYBERNETICS: SYSTEMS, Control Engineering Practice, Information Sciences, International Journal of Systems Science, and Neurocomputing.
