Implementation of analog ICs based on neural networks

Author

Smith, Michael J S

Author_Institution

Hawaii Univ., Honolulu, HI, USA

fYear

1989

fDate

14-16 Aug 1989

Firstpage

225

Abstract

Concepts found in the study of neural networks can be used to aid the analysis and increase the understanding of conventional analog ICs, as well as suggest new circuits and applications. The development of a series of analog ICs based on crossbar neural networks is presented. There are two main problems in their implementation: the choice of the correct weights to ensure that stable states correspond to solutions to the problem addressed by the network and the stability of the network which depends critically on its integrated circuit construction. The transfer curve of an analog IC implementation of an A/D converter illustrates the problems of choosing the weights in such networks correctly in order to avoid incorrect solutions

Keywords

analogue circuits; analogue-digital conversion; linear integrated circuits; neural nets; A/D converter; analog ICs; crossbar neural networks; integrated circuit construction; neural networks; stability; stable states; transfer curve; weights; Analog integrated circuits; CMOS analog integrated circuits; CMOS integrated circuits; Circuits and systems; Integrated circuit modeling; Logistics; Neural networks; Operational amplifiers; Servomechanisms; Temperature;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1989., Proceedings of the 32nd Midwest Symposium on

Conference_Location

Champaign, IL

Type

conf

DOI

10.1109/MWSCAS.1989.101834

Filename

101834