One major challenge with undertaking research at a public comprehensive university is finding an area that resonates well with undergraduates and is practical and achievable with limited time and minimal resources. Although my formal research training focused on distributed and networked systems, including cluster, grid, and cloud computing environments, I have been working to transition my research activities into the areas of digital privacy and digital forensics and anti-forensics. I have chosen these areas due to student interest, positive student feedback, alignment with my teaching, and a strategic department objective to produce publications in fields related to computer security, in order to pursue an eventual NSA Centers of Academic Excellence designation. These areas are a new undertaking for me as of the end of 2019, so I’m starting from the ground floor with this work.
The theme of my initial exploration is that digital privacy is directly related to anti-forensics, in that the same techniques that one might apply in an anti-forensic situation can lead to increased privacy. For example, an intelligence agent in the field must be extremely careful about the quantity, type, and nature of artifacts generated, whether intentional or not. Failure to control artifacts could lead to discovery, compromise the mission objectives, and even result in injury or death. While the consequences of most privacy breaches are not nearly as dire, the same elements are present in that artifacts are collected, assembled, and mined to discover information that can be used in negative ways. Minimizing artifact creation, and carefully controlling access to artifacts that must be created, have the potential to improve digital privacy for both individuals and corporations.
My current work is therefore centered around determining the types of artifacts that are created in the day-to-day use of computing devices and implementing security controls that either restrict or minimize creation of those artifacts. Digital forensic techniques become important here, since forensic tools provide a mechanism for discovering and classifying artifacts. Some improvements to open source tools and techniques are likely required as part of the research process.
The Coastal Carolina University Cyberinfrastructure
My prior research work was centered around Cyberinfrastructure and development of middleware and systems to support scientific research and education. The central theme was that computing environments should be tailored to the needs of the user, as opposed to forcing the user to adapt to whatever environment is convenient to provide. I focused on delivering and managing systems for scientific computation, especially for atmospheric models, oceanographic applications, and experiential learning in the computing and marine sciences. The ultimate culmination of this work was the Coastal Carolina University Cyberinfrastructure (formerly the Cyberinfrastructure Project), which was designed to be a private cloud system. It would provide a graphical desktop environment in a web browser, allowing users to access a variety of interconnected resources, including High Performance Computing (HPC) clusters and student-accessible virtualization systems.
As of late 2019, I turned over the completed project to CCU Information Technology Services, where it is fully supported on an ongoing basis by a dedicated system administrator. This system was being used for research in the atmospheric and oceanic sciences, as well as for teaching in the marine science, computing sciences, chemistry, and engineering science disciplines. Eight years of work culminated in $119,150 in total grant support, of which $108,524 was provided by a National Science Foundation Major Research Instrument (MRI) grant. While my primary obligation on the MRI grant was to implement and manage the Cyberinfrastructure system, I did manage to coauthor three peer-reviewed publications related to this work (one on the Pulley system management tool we created, and two on computer science pedagogy that made use of the system).
Virtual Organization Clusters
For my doctoral dissertation, I focused on distributed systems, particularly on grid and cloud computing. This direction led me to design a new architecture for grid computing known as the Virtual Organization Cluster Model, which permits participating entities to move end-user computation into autonomically provisioned virtual containers without disrupting or substantially decreasing the performance of existing production grids. In addition to the completed and defended dissertation, this work yielded 3 journal articles and 10 refereed conference proceedings.
Sensor Networks and Remote Sensing
My undergraduate and early graduate school research focused on sensor networks and remote sensing systems for environmental applications. This work included the use of Berkeley Motes and TinyOS, yielding systems that were deployed in the Clemson Experimental Forest for ephemeral stream detection research related to water quality applications. I also worked with Weather Surveillance Radar - 1988 Doppler (WSR-88D) data to visualize and quantify total accumulated precipitation over a watershed. This environmentally-focused work yielded a journal article, a proceeding in a computing conference, and two presentations in domain science conferences. I also completed an undergraduate honors thesis and presented two posters at a university research forum.
My career total funding as of early 2020 totals $328,250. This includes $230,024 from the National Science Foundation (MRI grant and Graduate Research Fellowship Program funding), $90,228 from collaborative internal grants at Coastal Carolina University (course development, professional enhancement, and assessment), and $7,998 from the South Carolina Space Grant Consortium.