Pulse stream based CNN hardware implementation

Author

Colodro, F. ; Torralba, A. ; Carvaja, R.G. ; Franquelo, L.G.

Author_Institution

Dept. de Ingenieria Electron., Escuela Superior de Ingenieros, Sevilla, Spain

Volume

3

fYear

1998

Firstpage

211

Abstract

This paper presents the application of Pulse Stream Techniques (PSTs) to the hardware implementation of a Cellular Neural Network (CNN). The time differential equations of this network suggest that the dynamic of one neuron status can be emulated by adding discretized packets of charge to a capacitor. This task can be carried out by driving a current source with a pulse stream signal. The stochastic representation of internal signals reduces hardware complexity

Keywords

CMOS integrated circuits; cellular neural nets; mixed analogue-digital integrated circuits; neural chips; pulse width modulation; 1 micron; 5 V; ASIC; CMOS IC; CNN hardware implementation; PWM; cellular neural network; discretized charge packets; pulse stream based implementation; stochastic signal representation; time differential equations; Capacitors; Cellular neural networks; Circuits; Differential equations; Hardware; Neural networks; Neurons; Pulse width modulation; Space vector pulse width modulation; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1998. ISCAS '98. Proceedings of the 1998 IEEE International Symposium on

Conference_Location

Monterey, CA

Print_ISBN

0-7803-4455-3

Type

conf

DOI

10.1109/ISCAS.1998.703979

Filename

703979