Project Description
The FCT Team will initially research and develop algorithms from biometric research on potentially disruptive computing technologies, leveraging the PI’s current research in High-performance Reconfigurable Computing (HPRC) and High-performance Computing (HPC) accelerators. The FCT team will be graduate student-led and run with input from Dr. Smith. Specifically, undergraduates will work with graduate students in the research laboratory on code development, performance analysis, profiling, and testing with the goal of producing publishable results.
Initially, the undergraduate students will be involved in biometric related application code development for various accelerator (FPGA, GPU, Xeon-Phi, etc.) platforms. The biometric research focus will extend current efforts to develop robust facial and iris recognition algorithms for use in covert biometric identification applications. The FCT Team will specifically investigate the use of distributed algorithms for biometrics and the use of accelerator platforms to improve the performance of these algorithms.
Equipment Available to Team
The FCT Team will use the equipment in the Future Computing Technologies Laboratory as the basis for experiments. This Laboratory includes a Reconfigurable Computing Cluster, GPU cluster, Xeon-Phi workstation, and multiple FPGA development boards. Additional equipment is available via remote access through partnerships with other academic research laboratories. More info : http://www.parl.clemson.edu/~smithmc/research.html.
The FCT Team consists of 2 to 3 subgroups focusing on the following areas:
- Future Computing Technology research and application
- Analysis of biometric methods and models
- Use of FCT architectures to solve biometric research problems
Each of the subgroups will be lead by a graduate research associate working directly with the faculty mentor. Within each focus area, there are several issues that will be addressed by the group over the life of the FCT Team:
- Accelerator platform performance envelope and models
- Application porting and implementation (code development)
- Methods and algorithms for analyzing biometric sample quality
- Models to predict recognition performance
- Matching application needs to computing platform
- Implementation issues
- Performance tuning and analysis
Intellectual Merit
The research activities performed by the team and the requirement that the students organize and run the team with guidance from the graduate student leader(s) and faculty will help achieve the following educational outcomes. Upon completion of a two-year tenure with the team, each student will have demonstrated :
- An ability to apply knowledge of mathematics, science, and engineering
- An ability to design and conduct experiments as well as analyze and interpret data
- An ability to design a system, component, or process to meet desired need within realistic constraints such as economic, environmental, social, political, ethical, health and safety, manufacturability, and sustainability
- An ability to function on multidisciplinary teams
- An ability to identify, formulate, and solve engineering problems
- Understanding of professional and ethical responsibility
- An ability to communicate effectively
- A broad education necessary to understand impact of engineering solutions in global, economic, environmental, and societal context
- An ability to use techniques, skills and modern engineering tools necessary for engineering practice