Design and validation of an artificial neural network based on analog circuits

Fikret Başar Gencer; Xhesila Xhafa; Benan Beril İnam; Mustafa Berke Yelten

doi:10.1007/s10470-020-01713-x

Design and validation of an artificial neural network based on analog circuits

Fikret Başar Gencer, Xhesila Xhafa, Benan Beril İnam, Mustafa Berke Yelten^*

^*Corresponding author for this work

Department of Electronics and Communication Engineering

Istanbul Technical University

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

This paper focuses on the design and validation of an analog artificial neural network. Basic building blocks of the analog ANN have been constructed in UMC 90 nm device technology. Performance metrics of the building blocks have been demonstrated through circuit simulations. The weights of the ANN have been estimated through an automated back-propagation algorithm, which is running circuit simulations during weight optimization. Two case studies, the operation an XOR logic gate and a full adder circuit have been captured using the proposed analog ANN. Monte Carlo analysis of the XOR gate reveals that the analog ANN operates with an accuracy of 99.85%.

Original language	English
Pages (from-to)	475-483
Number of pages	9
Journal	Analog Integrated Circuits and Signal Processing
Volume	106
Issue number	3
DOIs	https://doi.org/10.1007/s10470-020-01713-x
Publication status	Published - Mar 2021

Bibliographical note

Publisher Copyright:
© 2020, Springer Science+Business Media, LLC, part of Springer Nature.

Keywords

Analog circuits
Artificial neural networks
Full adder
XOR gate

Access to Document

10.1007/s10470-020-01713-x

Cite this

@article{f67cfd6e3afb430caa75b1d87a878801,

title = "Design and validation of an artificial neural network based on analog circuits",

abstract = "This paper focuses on the design and validation of an analog artificial neural network. Basic building blocks of the analog ANN have been constructed in UMC 90 nm device technology. Performance metrics of the building blocks have been demonstrated through circuit simulations. The weights of the ANN have been estimated through an automated back-propagation algorithm, which is running circuit simulations during weight optimization. Two case studies, the operation an XOR logic gate and a full adder circuit have been captured using the proposed analog ANN. Monte Carlo analysis of the XOR gate reveals that the analog ANN operates with an accuracy of 99.85%.",

keywords = "Analog circuits, Artificial neural networks, Full adder, XOR gate",

author = "Gencer, {Fikret Ba{\c s}ar} and Xhesila Xhafa and İnam, {Benan Beril} and Yelten, {Mustafa Berke}",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Science+Business Media, LLC, part of Springer Nature.",

year = "2021",

month = mar,

doi = "10.1007/s10470-020-01713-x",

language = "English",

volume = "106",

pages = "475--483",

journal = "Analog Integrated Circuits and Signal Processing",

issn = "0925-1030",

publisher = "Springer Netherlands",

number = "3",

}

TY - JOUR

T1 - Design and validation of an artificial neural network based on analog circuits

AU - Gencer, Fikret Başar

AU - Xhafa, Xhesila

AU - İnam, Benan Beril

AU - Yelten, Mustafa Berke

PY - 2021/3

Y1 - 2021/3

N2 - This paper focuses on the design and validation of an analog artificial neural network. Basic building blocks of the analog ANN have been constructed in UMC 90 nm device technology. Performance metrics of the building blocks have been demonstrated through circuit simulations. The weights of the ANN have been estimated through an automated back-propagation algorithm, which is running circuit simulations during weight optimization. Two case studies, the operation an XOR logic gate and a full adder circuit have been captured using the proposed analog ANN. Monte Carlo analysis of the XOR gate reveals that the analog ANN operates with an accuracy of 99.85%.

AB - This paper focuses on the design and validation of an analog artificial neural network. Basic building blocks of the analog ANN have been constructed in UMC 90 nm device technology. Performance metrics of the building blocks have been demonstrated through circuit simulations. The weights of the ANN have been estimated through an automated back-propagation algorithm, which is running circuit simulations during weight optimization. Two case studies, the operation an XOR logic gate and a full adder circuit have been captured using the proposed analog ANN. Monte Carlo analysis of the XOR gate reveals that the analog ANN operates with an accuracy of 99.85%.

KW - Analog circuits

KW - Artificial neural networks

KW - Full adder

KW - XOR gate

UR - http://www.scopus.com/inward/record.url?scp=85091110555&partnerID=8YFLogxK

U2 - 10.1007/s10470-020-01713-x

DO - 10.1007/s10470-020-01713-x

M3 - Article

AN - SCOPUS:85091110555

SN - 0925-1030

VL - 106

SP - 475

EP - 483

JO - Analog Integrated Circuits and Signal Processing

JF - Analog Integrated Circuits and Signal Processing

IS - 3

ER -

Design and validation of an artificial neural network based on analog circuits

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this