DocumentCode :
358269
Title :
The CNNUC3: an analog I/O 64x64 CNN universal machine chip prototype with 7-bit analog accuracy
Author :
Linán, G. ; Espejo, S. ; Domìnguez-Castro, R. ; Rodriguez-Vazquen, A.
Author_Institution :
Inst. de Microelectron., CSIC, Sevilla, Spain
fYear :
2000
fDate :
2000
Firstpage :
201
Lastpage :
206
Abstract :
This paper describes a full-custom mixed-signal chip which embeds distributed optical signal acquisition, digitally-programmable analog parallel processing, and distributed image memory (cache) on a common silicon substrate. This chip, designed in a 0.5 μm CMOS standard technology contains around 1000000 transistors, 80% of which operate in analog mode. Chip functional features are in accordance to the CNN Universal Machine paradigm. The chip is capable to complete complex spatio-temporal image processing tasks within short computation time and using a low power budget. The internal circuitry of the chip has been designed to operate in robust manner with >7-bit equivalent accuracy in the internal analog operations, which has been confirmed by experimental measurements. Hence, to all practical purposes, processing tasks completed by the chip have the same accuracy than those completed by digital processors preceded by 7-bit digital-to-analog converters for image digitalization
Keywords :
CMOS integrated circuits; cellular neural nets; image processing; mixed analogue-digital integrated circuits; neural chips; parallel processing; CMOS; CNNUC3; analog parallel processing; cellular neural networks; distributed image memory; image processing; mixed-signal chip; neural chips; universal machine chip; CMOS technology; Cellular neural networks; Circuits; Image processing; Parallel processing; Robustness; Semiconductor device measurement; Signal processing; Silicon; Turing machines;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Cellular Neural Networks and Their Applications, 2000. (CNNA 2000). Proceedings of the 2000 6th IEEE International Workshop on
Conference_Location :
Catania
Print_ISBN :
0-7803-6344-2
Type :
conf
DOI :
10.1109/CNNA.2000.876845
Filename :
876845
Link To Document :
بازگشت