A programmable processor for the Cheops image processing system
Year: 1995
Type: dissertation
Abstract: We explore the use of dynamically reconfigurable hardware in the CHEOPS image processing system.CHEOPS incurs a substantial performance degradation when executing computational operations for which no dedicated stream processor exists.A new programmable processor that combines the speed of special purpose stream architectures with the flexibility of general purpose processing is proposed and implemented as a solution to this problem.Two SRAM based FPGAs are utilized in conjunction with a PC603 microprocessor to provide a flexible computational substrate.The system allows algorithms to be arbitrarily mapped to some combination of dedicated hardware, and software within the data flow paradigm.Dynamic reconfiguration allows the processor to specialize to each computation while maintaining temporal flexibility.The system is designed and implemented; observations are reported.
Author: Edward Kelly Acosta
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: MIT Media Lab
OpenAlex ID: W1603385004
