Knight Foundation School of Computing and Information Sciences
Yuzhou Feng is a Ph.D. candidate in the Knight Foundation School of Computing and Information Sciences (KFSCIS) at Florida International University (FIU). He is currently working in the Cyber-Physical System Laboratory (CPSLab) under the supervision of Dr. Dong Chen. He received his first M.S. in Interaction Design from Beihang University, China, and the second M.S. in Engineering Management of Computer Science from FIU. His research mainly focuses on building data analytics systems with a particular emphasis on system sustainability and data privacy of CPS application systems, such as smart homes, smart grids, and smart cities. He has published papers at top-tier sensor system conferences, such as ACM/IEEE IPSN 2020, ACM BuildSys 2020. His most recent work SolarTrader won the Best Paper Award at ACM BuildSys 2020.
The cyber-physical system (CPS) has been erupting worldwide over the past decade. People in smart homes and smart buildings are increasingly deploying Internet of Things(IoT) devices to monitor and control their environments; moreover, distributed solar energy resources (DSERs) in smart grid systems are rapidly increasing due to the steep decline in solar module prices. In each area, new IoT-enabled devices are rapidly developed. The total installed base of the IoT-connected devices is projected to amount to 75.44 billion worldwide by 2025, a fivefold increase in ten years. To accommodate big data-driven CPS system decisions, there are multiple challenges that we need to handle, such as fundamental understanding of data analytics modeling, secure system architecture, security, and privacy strengthening techniques.
In order to address these challenges, this research is focusing on how to leverage data analytic approaches, such as machine learning, deep learning, and reinforcement learning approaches, to analyze the big data generated by the CPS to make it more efficient in a secure manner. This research mainly focuses on leveraging data analytic approaches, such as machine learning, deep learning, and reinforcement learning techniques, on analyzing the big data generated by the CPS to make it more efficient in a secure manner. The proposed work includes various data frameworks in CPS fields: 1) How to build a framework to detect vulnerability to multiple cyberattacks and how to use network traffic traces inferring serious user in-home private information leakage? 2) How to enable fairness-aware energy trading in DSERs environment? 3) How to build an integrated CPS data analytics framework that users can visualize and manage CPS data in real-time?