Title :
A neural network systems component
Author :
Mueller, Dean ; Hammerstrom, Dan
Author_Institution :
Adaptive Solutions Inc., Beaverton, OR, USA
Abstract :
A CMOS device designed to interface the CNAPS (connected network of adaptive processors)-1064 neural processor IC to traditional computing environments is presented. The encapsulation of flexible I/O models and sequencer control into a modular architecture enables the creation of systems across a wide spectrum of performance capabilities. The CNAPS sequencer chip (CSC) provides a simple co-processor interface that can be easily adapted to different target environments and, at the same time, utilize the CNAPS architecture at the maximum performance possible
Keywords :
CMOS integrated circuits; computer interfaces; microprocessor chips; neural chips; CMOS device; CNAPS; adaptive processors; co-processor interface; connected network; flexible I/O models; modular architecture; neural processor IC; performance capabilities; sequencer control; target environments; Adaptive systems; CMOS integrated circuits; CMOS process; Computer architecture; Computer interfaces; Computer networks; Coprocessors; Encapsulation; Neural networks; Semiconductor device modeling;
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
DOI :
10.1109/ICNN.1993.298738
