An Energy-Efficient Solid-State Organic Device Array for Neuromorphic Computing

Lan Shen Hu; Marco Fattori; Winston Schilp; Roy Verbeek; Setareh Kazemzadeh; Yoeri van de Burgt; Auke Jisk Kronemeijer; Gerwin Gelinck; Eugenio Cantatore

doi:10.1109/TED.2023.3327947

An Energy-Efficient Solid-State Organic Device Array for Neuromorphic Computing

Lan Shen Hu (Corresponding author), Marco Fattori, Winston Schilp, Roy Verbeek, Setareh Kazemzadeh, Yoeri van de Burgt, Auke Jisk Kronemeijer, Gerwin Gelinck, Eugenio Cantatore

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

3 Citations (Scopus)

13 Downloads (Pure)

Abstract

The slowing-down of Moore’s law is shifting the computing paradigm towards solutions based on quantum and neuromorphic computing elements. Unlike conventional digital computing, neuromorphic computing is based on analog devices. In this work, we propose a three-terminal neuromorphic organic device (NODe) capable of providing both analog computing and memory in a single device by tuning its conductance. The availability of three-terminal devices enables the independent tuning of the NODes, preventing write sneak path issues typical of the two-terminal memristor crossbar array. The NODe conductance relaxes exponentially with a measured time constant of 2.9 h, furthermore, it can be operated at 51 MHz, corresponding to an estimated energy efficiency of 0.1 pJ per multiply-accumulate (MAC) operation. To demonstrate the NODe’s computing capabilities, a 3×3 crossbar array has been successfully used to perform edge detection and blurring on an image with 128×64 pixels.

Original language	English
Pages (from-to)	6520-6525
Number of pages	6
Journal	IEEE Transactions on Electron Devices
Volume	70
Issue number	12
DOIs	https://doi.org/10.1109/TED.2023.3327947
Publication status	Published - 1 Dec 2023

Bibliographical note

Publisher Copyright:
IEEE

Keywords

Artificial neuron
Capacitance
Capacitors
crossbar array
energy efficient
Integrated circuit modeling
Logic gates
neuromorphic
parallel computing
Programming
Tuning
Voltage measurement

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/TED.2023.3327947

pdfFinal published version, 6.53 MBLicence: TAVERNE

Cite this

@article{5a740ded794c47afb7bb71ea6359d851,

title = "An Energy-Efficient Solid-State Organic Device Array for Neuromorphic Computing",

abstract = "The slowing-down of Moore{\textquoteright}s law is shifting the computing paradigm towards solutions based on quantum and neuromorphic computing elements. Unlike conventional digital computing, neuromorphic computing is based on analog devices. In this work, we propose a three-terminal neuromorphic organic device (NODe) capable of providing both analog computing and memory in a single device by tuning its conductance. The availability of three-terminal devices enables the independent tuning of the NODes, preventing write sneak path issues typical of the two-terminal memristor crossbar array. The NODe conductance relaxes exponentially with a measured time constant of 2.9 h, furthermore, it can be operated at 51 MHz, corresponding to an estimated energy efficiency of 0.1 pJ per multiply-accumulate (MAC) operation. To demonstrate the NODe{\textquoteright}s computing capabilities, a 3×3 crossbar array has been successfully used to perform edge detection and blurring on an image with 128×64 pixels.",

keywords = "Artificial neuron, Capacitance, Capacitors, crossbar array, energy efficient, Integrated circuit modeling, Logic gates, neuromorphic, parallel computing, Programming, Tuning, Voltage measurement",

author = "Hu, {Lan Shen} and Marco Fattori and Winston Schilp and Roy Verbeek and Setareh Kazemzadeh and {van de Burgt}, Yoeri and Kronemeijer, {Auke Jisk} and Gerwin Gelinck and Eugenio Cantatore",

note = "Publisher Copyright: IEEE",

year = "2023",

month = dec,

day = "1",

doi = "10.1109/TED.2023.3327947",

language = "English",

volume = "70",

pages = "6520--6525",

journal = "IEEE Transactions on Electron Devices",

issn = "0018-9383",

publisher = "Institute of Electrical and Electronics Engineers",

number = "12",

}

TY - JOUR

T1 - An Energy-Efficient Solid-State Organic Device Array for Neuromorphic Computing

AU - Hu, Lan Shen

AU - Fattori, Marco

AU - Schilp, Winston

AU - Verbeek, Roy

AU - Kazemzadeh, Setareh

AU - van de Burgt, Yoeri

AU - Kronemeijer, Auke Jisk

AU - Gelinck, Gerwin

AU - Cantatore, Eugenio

N1 - Publisher Copyright: IEEE

PY - 2023/12/1

Y1 - 2023/12/1

N2 - The slowing-down of Moore’s law is shifting the computing paradigm towards solutions based on quantum and neuromorphic computing elements. Unlike conventional digital computing, neuromorphic computing is based on analog devices. In this work, we propose a three-terminal neuromorphic organic device (NODe) capable of providing both analog computing and memory in a single device by tuning its conductance. The availability of three-terminal devices enables the independent tuning of the NODes, preventing write sneak path issues typical of the two-terminal memristor crossbar array. The NODe conductance relaxes exponentially with a measured time constant of 2.9 h, furthermore, it can be operated at 51 MHz, corresponding to an estimated energy efficiency of 0.1 pJ per multiply-accumulate (MAC) operation. To demonstrate the NODe’s computing capabilities, a 3×3 crossbar array has been successfully used to perform edge detection and blurring on an image with 128×64 pixels.

AB - The slowing-down of Moore’s law is shifting the computing paradigm towards solutions based on quantum and neuromorphic computing elements. Unlike conventional digital computing, neuromorphic computing is based on analog devices. In this work, we propose a three-terminal neuromorphic organic device (NODe) capable of providing both analog computing and memory in a single device by tuning its conductance. The availability of three-terminal devices enables the independent tuning of the NODes, preventing write sneak path issues typical of the two-terminal memristor crossbar array. The NODe conductance relaxes exponentially with a measured time constant of 2.9 h, furthermore, it can be operated at 51 MHz, corresponding to an estimated energy efficiency of 0.1 pJ per multiply-accumulate (MAC) operation. To demonstrate the NODe’s computing capabilities, a 3×3 crossbar array has been successfully used to perform edge detection and blurring on an image with 128×64 pixels.

KW - Artificial neuron

KW - Capacitance

KW - Capacitors

KW - crossbar array

KW - energy efficient

KW - Integrated circuit modeling

KW - Logic gates

KW - neuromorphic

KW - parallel computing

KW - Programming

KW - Tuning

KW - Voltage measurement

UR - http://www.scopus.com/inward/record.url?scp=85177056329&partnerID=8YFLogxK

U2 - 10.1109/TED.2023.3327947

DO - 10.1109/TED.2023.3327947

M3 - Article

AN - SCOPUS:85177056329

SN - 0018-9383

VL - 70

SP - 6520

EP - 6525

JO - IEEE Transactions on Electron Devices

JF - IEEE Transactions on Electron Devices

IS - 12

ER -

An Energy-Efficient Solid-State Organic Device Array for Neuromorphic Computing

Abstract

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this