A `neural´ A/D converter utilizing Schmitt trigger

Author

Cataldo, G. Di ; Palumbo, G.

Author_Institution

Istituto di Elettrotecnica & Electtronica, Catania Univ., Italy

fYear

1990

fDate

1-3 May 1990

Firstpage

1078

Abstract

Theoretical considerations are presented regarding the stability and the physical realizability of a new A/D converter based on the Hopfield neural network. The equivalent circuit made and described by M.J.S. Smith and C.L. Portmann (1989) is analyzed in order to solve the problem in a different way. The authors adopt the same circuit scheme realized with Schmitt triggers, but modifying their hysteresis and thresholds in an appropriate manner. The role played by the amplifier response time is widely illustrated, and the network theoretical behavior is analyzed. A proof that validates the correct operation of the circuits is reported

Keywords

analogue-digital conversion; neural nets; switching networks; trigger circuits; ADC; Hopfield neural network; Schmitt trigger; amplifier response time; correct operation; equivalent circuit; hysteresis; network theoretical behavior; neural network A/D converter; physical realizability; thresholds; Computer networks; Convergence; Delay; Hopfield neural networks; Hysteresis; Limit-cycles; Neural networks; Neurons; Time factors; Trigger circuits;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1990., IEEE International Symposium on

Conference_Location

New Orleans, LA

Type

conf

DOI

10.1109/ISCAS.1990.112297

Filename

112297