Cyber-Physical Security addresses the security issues of cyber-physical integrated systems. Deep embedment of digital technology into every corner of modern society introduces strong correlation between physical and cyber activities. Through statistical inference, a community interconnected by private channels may not be able to conceal their organization and activities. We have both static and dynamic modeling efforts in this area. The research calls for multiple skill sets, ranging from data packet filtering, distributed systems, statistics, mining algorithms, and human behavior modeling. Applications range from situational awareness of tactical operations in traditional warfare, insider detection, disaster response, and crowd control. It also finds applications in measuring responsiveness of mission teams. We welcome interested parties to collaborate on this challenging research area to solve real world problems. (2010-current)
Global Positioning System (GPS) is one of the most important modern technologies to provide time and positioning information. Timed RF signals emitted from precisely calculated positions form an interlocked temporal-spatial grid for GPS receivers to calculate their own positions. We are working on modeling of the interlock relationship between GPS signals to establish necessary and sufficient conditions to assert signal integrity. One application of such a model is for detection of anomalous (spoofing and jamming) signals under different conditions. Another line of new efforts is related to development of synchronous computing architectures for advanced applications. (2010-current)
Screening and wellbeing Technology is a long standing effort to improve healthcare at reduced cost. We have developed a retinal laser injury detection algorithm (deployed), a portable medical information management system (deployed), a tool suite for remote image acquisition and upload management, and a lab version of batch processing of retinal images. The foundation of these artifacts is based on modeling and mapping algorithms for blood vessels in retinal images. We now look into correlated screening methods by taking advantage of miniaturized sensors and smart phones. One particular direction is to develop a crowd-sourced architecture to detect emerging health conditions based on a subject’s best fit demographic profile. (2002-present)
Past Projects
Biomedical Modeling kicked off a data driven, pattern analysis direction to meet the needs in both the cyber and physical worlds. We developed a proof process of a clustered structure on the transport of cholesterol over cell membrane. In a separate effort, we developed an algorithm suite to process aorta images, in order to generate distributions of lamina layer thickness. We have also worked on wavelet based image compression, image processing techniques, including a wavelet toolware to meet education needs at infancy of the wavelet research. (1994-2002)
Intelligent Transportation Systems (ITS) marked our first step in the interdisciplinary research work. Since inception of the ITS community, we have worked with the ITS R&D community to advocate distributed computing algorithms, networking protocols, and sensor data fusion at a time when most traffic signals worked in isolation with little or no networking infrastructure available. We developed an Internet based query system for ITS data (DataLink), which was integrated into the TransLink Lab of Texas Transportation Institute. (1989-2003)