Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications

Aijaz H. Lone; Hanrui Li; Nazek El-Atab; Gianluca Setti; Hossein Fariborzi

doi:10.1109/NANO58406.2023.10231303

Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications

Aijaz H. Lone, Hanrui Li, Nazek El-Atab, Gianluca Setti, Hossein Fariborzi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

We propose a novel spin-orbit torque (SOT) driven and voltage-gated domain wall motion (DWM)-based MTJ device and its application in neuromorphic computing. We show that by utilizing the voltage-controlled gating effect on the DWM, the access transistor can be eliminated. The device provides more control over individual synapse writing and shows highly linear synaptic behavior. The voltage-controlled DW- gating performance of the device in the crossbar array is quantified. The device long-term potentiation/depression (LTP/LTD) linearity dependence on material parameters such as DMI, at different temperatures, is evaluated for real-environment performance analysis. Furthermore, adopting the ideal and skyrmion leaky integrate and fire neuron models, we implement the spiking convolutional neural network for pattern recognition applications The DW-MTJ conductance was mapped to the weights of the SNN. When trained and tested on the CIFAR-10 data set, the architecture based on DW-MTJ synapse achieved accuracy of around 85%.

Original language	English (US)
Title of host publication	2023 IEEE 23rd International Conference on Nanotechnology, NANO 2023
Publisher	IEEE Computer Society
Pages	976-981
Number of pages	6
ISBN (Electronic)	9798350333466
DOIs	https://doi.org/10.1109/NANO58406.2023.10231303
State	Published - 2023
Event	23rd IEEE International Conference on Nanotechnology, NANO 2023 - Jeju City, Korea, Republic of Duration: Jul 2 2023 → Jul 5 2023

Publication series

Name	Proceedings of the IEEE Conference on Nanotechnology
Volume	2023-July
ISSN (Print)	1944-9399
ISSN (Electronic)	1944-9380

Conference

Conference	23rd IEEE International Conference on Nanotechnology, NANO 2023
Country/Territory	Korea, Republic of
City	Jeju City
Period	07/2/23 → 07/5/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

Domain wall devices
MTJ
Neuromorphic computing
Skyrmions
SNN
Spintronics
Synapses
VCMA

ASJC Scopus subject areas

Bioengineering
Electrical and Electronic Engineering
Materials Chemistry
Condensed Matter Physics

Access to Document

10.1109/NANO58406.2023.10231303

Cite this

Lone, A. H., Li, H., El-Atab, N., Setti, G., & Fariborzi, H. (2023). Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications. In 2023 IEEE 23rd International Conference on Nanotechnology, NANO 2023 (pp. 976-981). (Proceedings of the IEEE Conference on Nanotechnology; Vol. 2023-July). IEEE Computer Society. https://doi.org/10.1109/NANO58406.2023.10231303

@inproceedings{f05656a0583a460c9db8092a3055d23f,

title = "Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications",

abstract = "We propose a novel spin-orbit torque (SOT) driven and voltage-gated domain wall motion (DWM)-based MTJ device and its application in neuromorphic computing. We show that by utilizing the voltage-controlled gating effect on the DWM, the access transistor can be eliminated. The device provides more control over individual synapse writing and shows highly linear synaptic behavior. The voltage-controlled DW- gating performance of the device in the crossbar array is quantified. The device long-term potentiation/depression (LTP/LTD) linearity dependence on material parameters such as DMI, at different temperatures, is evaluated for real-environment performance analysis. Furthermore, adopting the ideal and skyrmion leaky integrate and fire neuron models, we implement the spiking convolutional neural network for pattern recognition applications The DW-MTJ conductance was mapped to the weights of the SNN. When trained and tested on the CIFAR-10 data set, the architecture based on DW-MTJ synapse achieved accuracy of around 85%.",

keywords = "Domain wall devices, MTJ, Neuromorphic computing, Skyrmions, SNN, Spintronics, Synapses, VCMA",

author = "Lone, {Aijaz H.} and Hanrui Li and Nazek El-Atab and Gianluca Setti and Hossein Fariborzi",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 23rd IEEE International Conference on Nanotechnology, NANO 2023 ; Conference date: 02-07-2023 Through 05-07-2023",

year = "2023",

doi = "10.1109/NANO58406.2023.10231303",

language = "English (US)",

series = "Proceedings of the IEEE Conference on Nanotechnology",

publisher = "IEEE Computer Society",

pages = "976--981",

booktitle = "2023 IEEE 23rd International Conference on Nanotechnology, NANO 2023",

address = "United States",

}

Lone, AH , Li, H , El-Atab, N , Setti, G & Fariborzi, H 2023, Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications. in 2023 IEEE 23rd International Conference on Nanotechnology, NANO 2023. Proceedings of the IEEE Conference on Nanotechnology, vol. 2023-July, IEEE Computer Society, pp. 976-981, 23rd IEEE International Conference on Nanotechnology, NANO 2023, Jeju City, Korea, Republic of, 07/2/23. https://doi.org/10.1109/NANO58406.2023.10231303

Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications. / Lone, Aijaz H.; Li, Hanrui ; El-Atab, Nazek et al.
2023 IEEE 23rd International Conference on Nanotechnology, NANO 2023. IEEE Computer Society, 2023. p. 976-981 (Proceedings of the IEEE Conference on Nanotechnology; Vol. 2023-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications

AU - Lone, Aijaz H.

AU - Li, Hanrui

AU - El-Atab, Nazek

AU - Setti, Gianluca

AU - Fariborzi, Hossein

PY - 2023

Y1 - 2023

N2 - We propose a novel spin-orbit torque (SOT) driven and voltage-gated domain wall motion (DWM)-based MTJ device and its application in neuromorphic computing. We show that by utilizing the voltage-controlled gating effect on the DWM, the access transistor can be eliminated. The device provides more control over individual synapse writing and shows highly linear synaptic behavior. The voltage-controlled DW- gating performance of the device in the crossbar array is quantified. The device long-term potentiation/depression (LTP/LTD) linearity dependence on material parameters such as DMI, at different temperatures, is evaluated for real-environment performance analysis. Furthermore, adopting the ideal and skyrmion leaky integrate and fire neuron models, we implement the spiking convolutional neural network for pattern recognition applications The DW-MTJ conductance was mapped to the weights of the SNN. When trained and tested on the CIFAR-10 data set, the architecture based on DW-MTJ synapse achieved accuracy of around 85%.

AB - We propose a novel spin-orbit torque (SOT) driven and voltage-gated domain wall motion (DWM)-based MTJ device and its application in neuromorphic computing. We show that by utilizing the voltage-controlled gating effect on the DWM, the access transistor can be eliminated. The device provides more control over individual synapse writing and shows highly linear synaptic behavior. The voltage-controlled DW- gating performance of the device in the crossbar array is quantified. The device long-term potentiation/depression (LTP/LTD) linearity dependence on material parameters such as DMI, at different temperatures, is evaluated for real-environment performance analysis. Furthermore, adopting the ideal and skyrmion leaky integrate and fire neuron models, we implement the spiking convolutional neural network for pattern recognition applications The DW-MTJ conductance was mapped to the weights of the SNN. When trained and tested on the CIFAR-10 data set, the architecture based on DW-MTJ synapse achieved accuracy of around 85%.

KW - Domain wall devices

KW - MTJ

KW - Neuromorphic computing

KW - Skyrmions

KW - SNN

KW - Spintronics

KW - Synapses

KW - VCMA

UR - http://www.scopus.com/inward/record.url?scp=85173591900&partnerID=8YFLogxK

U2 - 10.1109/NANO58406.2023.10231303

DO - 10.1109/NANO58406.2023.10231303

M3 - Conference contribution

AN - SCOPUS:85173591900

T3 - Proceedings of the IEEE Conference on Nanotechnology

SP - 976

EP - 981

BT - 2023 IEEE 23rd International Conference on Nanotechnology, NANO 2023

PB - IEEE Computer Society

T2 - 23rd IEEE International Conference on Nanotechnology, NANO 2023

Y2 - 2 July 2023 through 5 July 2023

ER -

Voltage Gated Domain Wall Magnetic Tunnel Junction for Neuromorphic Computing Applications

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this