The authors propose a new CAD methodology for highly parallel signal processors. The main issue is how to express parallel algorithms in a notation which is easy to understand by humans and possible to compile into efficient VLSI circuits and/or array processor (systolic/wavefront) machine codes. More specifically, they propose a methodology based on a hierarchical and recursive mapping from algorithms to VLSI array hardwares. In the first phase of the design a powerful signal flow graph (SFG) notation is used to express recurrence and ordering associated with the description of the space-time activities. In the second phase, this description is mapped into a VLSI hardware description or in executable array processor machine codes.
|Original language||English (US)|
|Number of pages||4|
|Journal||ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings|
|State||Published - Dec 1 1985|
All Science Journal Classification (ASJC) codes
- Signal Processing
- Electrical and Electronic Engineering