Thesis
CAD tool emulation for a two-level reconfigurable cell array for digital signal processing
Master of Science (MS), Washington State University
2005
Handle:
https://hdl.handle.net/2376/419
Abstract
The use of reconfigurable hardware has been increasing in recent years; the need for robust Computer Aided Design (CAD) tools has risen accordingly. For a reconfigurable system, CAD tools enable developers to create, modify, simulate, and estimate the performance of synthesized designs. Furthermore, once a design is realized, the CAD tools can be used to generate a configuration file to program the hardware. This research deals with the design of comprehensive CAD tools for a medium-grain reconfigurable cell array. This architecture has been developed by the High Performance Computer Systems (HiPerCopS) research group at Washington State University to accelerate digital signal processing (DSP). The CAD tools include a full-featured designer utility that allows users to map sophisticated DSP algorithms onto the array. A hardware-level simulator verifies these designs and produces results for benchmarking. Finally, an array of supporting structures, such as designer history and part libraries, are built around the core tools to provide a full-featured user experience. The CAD tools have been used to evaluate the reconfigurable architecture itself. The results from the simulated benchmarks help verify the performance and functionality of the system, and suggest potential areas for improvement. As a result, a more powerful interconnection network has been designed to simplify the mapping process and to improve performance. Thus, the CAD tools provide a powerful platform for future research into medium-grain reconfigurable architectures.
Metrics
1 File views/ downloads
14 Record Views
Details
- Title
- CAD tool emulation for a two-level reconfigurable cell array for digital signal processing
- Creators
- Jonathan Karl Larson
- Contributors
- José G. Delgado-Frias (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Identifiers
- 99900525286001842
- Language
- English
- Resource Type
- Thesis