An asynchronous pulse neural network model and its analog IC implementation

Author

Horio, Yoshihiko ; Yasuda, Hidekazu ; Hanagata, Mitsuru ; Aihara, Kazuyuki

Author_Institution

Dept. of Electron. Eng., Tokyo Denki Univ., Japan

Volume

3

fYear

1998

fDate

1998

Firstpage

301

Abstract

Prototype neuron and axon circuits for the asynchronous pulse neural network model are fabricated with 1.2 μm CMOS technology and tested. The neuron circuit can function as both of a coincidence detector and an integrator. These opposite neuronal functions can be controlled by changing the internal time constants through the external voltages. The axon circuit, which mimics axonal, dendritic and synaptic delays, realizes pulse delay. The delay time can be continuously controlled by the external voltage

Keywords

CMOS analogue integrated circuits; asynchronous circuits; coincidence circuits; delays; integrating circuits; neural nets; 1.2 mum; CMOS; asynchronous pulse neural network model; axon circuits; axonal delays; coincidence detector; delay time; dendritic delays; external voltages; integrator; internal time constants; neuronal functions; prototype neuron circuits; pulse delay; synaptic delays; Analog integrated circuits; CMOS analog integrated circuits; CMOS technology; Delay; Integrated circuit modeling; Nerve fibers; Neural networks; Neurons; Pulse circuits; Semiconductor device modeling;

fLanguage

English

Publisher

ieee

Conference_Titel

Electronics, Circuits and Systems, 1998 IEEE International Conference on

Conference_Location

Lisboa

Print_ISBN

0-7803-5008-1

Type

conf

DOI

10.1109/ICECS.1998.813995

Filename

813995