An Analogue CMOS Neural Circuit for Improved Sensing

Author

Zatorre, Uillermo ; Sanz, M. Teresa ; Medrano, Nicolas ; Martinez, Pedro A. ; Celma, Santiago

Author_Institution

Fac. de Ciencias, Univ. de Zaragoza

fYear

0

fDate

0-0 0

Firstpage

185

Lastpage

188

Abstract

This paper presents the design and simulation of a conditioning circuit based on a CMOS current-mode digitally programmable analogue adaptive processor. The proposed processor model consists of two main blocks: A mixed-signal four-quadrant multiplier and a class AB current conveyor that implements the non-linear output function. Starting from circuit simulation results, the behaviour of the proposed processor was modelled. A multilayer perceptron network implemented with the new artificial neuron structure was trained to tackle linearization of a giant magneto-resistive sensor. Simulation results show the efficiency of the new implementation

Keywords

CMOS analogue integrated circuits; current conveyors; current-mode circuits; integrated circuit design; logic design; microprocessor chips; multilayer perceptrons; multiplying circuits; neural chips; programmable circuits; analogue CMOS neural circuit; class AB current conveyor; conditioning circuit; current-mode processor; digitally programmable analogue adaptive processor; giant magneto-resistive sensor; mixed-signal four-quadrant multiplier; multilayer perceptron network; nonlinear output function; Artificial neural networks; CMOS analog integrated circuits; CMOS process; Circuit simulation; Electronic circuits; Magnetic sensors; Multilayer perceptrons; Neurons; Semiconductor device modeling; Transfer functions;

fLanguage

English

Publisher

ieee

Conference_Titel

Research in Microelectronics and Electronics 2006, Ph. D.

Conference_Location

Otranto

Print_ISBN

1-4244-0157-7

Type

conf

DOI

10.1109/RME.2006.1689927

Filename

1689927