Design of neural network systems from custom analog VLSI chips

Author

Wang, Yiwen ; Salam, Fathi M A

Author_Institution

Dept. of Electr. Eng., Michigan State Univ., East Lansing, MI, USA

fYear

1990

fDate

1-3 May 1990

Firstpage

1098

Abstract

Custom analog CMOS VLSI components have been designed to allow the construction of neural networks with arbitrary architectures. A wide-range, all-enhancement-mode, MOS analog four-quadrant multiplier has been employed to implement the scalar vector product of the vector of neuron outputs and the vector of the corresponding weights. A nonlinear MOSFET floating element is used to model nonlinear conductive elements; this extends the modeling of the synaptic weights to nonlinear elements. A neuron is realized by a simple CMOS operational amplifier which is compatible with the I/O of the analog multiplier. The neural system can be expanded modularly to large dimensions. SPICE simulations demonstrate the functionality of the all-MOS feedback neural nets using a two-neuron prototype as an example

Keywords

CMOS integrated circuits; VLSI; analogue computer circuits; application specific integrated circuits; linear integrated circuits; multiplying circuits; neural nets; operational amplifiers; CMOS; CMOS operational amplifier; SPICE simulations; all enhancement mode MOS multiplier; analog four-quadrant multiplier; analog multiplier; arbitrary architectures; construction of neural networks; custom analog VLSI chips; expanded modularly to large dimensions; feedback neural nets; model nonlinear conductive elements; neural network systems design; nonlinear MOSFET floating element; nonlinear elements; scalar vector product; synaptic weights; two-neuron prototype; vector of neuron outputs; Artificial neural networks; CMOS analog integrated circuits; MOSFET circuits; Neural networks; Neurons; Operational amplifiers; SPICE; Semiconductor device modeling; Vectors; Very large scale integration;

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.1990.112305

Filename

112305