Capacitor-Based Synaptic Devices for Hardware Spiking Neural Networks

Sungmin Hwang; Junsu Yu; Geun Ho Lee; Min Suk Song; Jeesoo Chang; Kyung Kyu Min; Taejin Jang; Jong Ho Lee; Byung Gook Park; Hyungjin Kim

doi:10.1109/LED.2022.3149029

Capacitor-Based Synaptic Devices for Hardware Spiking Neural Networks

Sungmin Hwang, Junsu Yu, Geun Ho Lee, Min Suk Song, Jeesoo Chang, Kyung Kyu Min, Taejin Jang, Jong Ho Lee, Byung Gook Park, Hyungjin Kim

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

27 Scopus citations

Abstract

In this work, we present a hardware neural network with capacitor-based synaptic devices. A capacitor-based synaptic device was developed using a MOS capacitor structure with a charge trapping layer. Due to the flat band voltage shift by charge trapping and its non-linear {C} - {V} characteristics, multilevel weight values could be implemented by the charge occurring when charging and discharging the capacitor. The vector-matrix multiplication (VMM) function was also experimentally verified using a fabricated synapse array based on NAND flash structure.

Original language	English
Pages (from-to)	549-552
Number of pages	4
Journal	IEEE Electron Device Letters
Volume	43
Issue number	4
DOIs	https://doi.org/10.1109/LED.2022.3149029
State	Published - 1 Apr 2022

Bibliographical note

Publisher Copyright:
© 1980-2012 IEEE.

Keywords

Capacitor-based synaptic device
MOS capacitor
NAND flash memory
capacitive neural network
neuromorphic system
spiking neural network (SNN)

Access to Document

10.1109/LED.2022.3149029

Cite this

@article{2a17847f6c164e8d8ca6e9966dc15789,

title = "Capacitor-Based Synaptic Devices for Hardware Spiking Neural Networks",

abstract = "In this work, we present a hardware neural network with capacitor-based synaptic devices. A capacitor-based synaptic device was developed using a MOS capacitor structure with a charge trapping layer. Due to the flat band voltage shift by charge trapping and its non-linear {C} - {V} characteristics, multilevel weight values could be implemented by the charge occurring when charging and discharging the capacitor. The vector-matrix multiplication (VMM) function was also experimentally verified using a fabricated synapse array based on NAND flash structure.",

keywords = "Capacitor-based synaptic device, MOS capacitor, NAND flash memory, capacitive neural network, neuromorphic system, spiking neural network (SNN)",

author = "Sungmin Hwang and Junsu Yu and Lee, {Geun Ho} and Song, {Min Suk} and Jeesoo Chang and Min, {Kyung Kyu} and Taejin Jang and Lee, {Jong Ho} and Park, {Byung Gook} and Hyungjin Kim",

note = "Publisher Copyright: {\textcopyright} 1980-2012 IEEE.",

year = "2022",

month = apr,

day = "1",

doi = "10.1109/LED.2022.3149029",

language = "English",

volume = "43",

pages = "549--552",

journal = "IEEE Electron Device Letters",

issn = "0741-3106",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "4",

}

TY - JOUR

T1 - Capacitor-Based Synaptic Devices for Hardware Spiking Neural Networks

AU - Hwang, Sungmin

AU - Yu, Junsu

AU - Lee, Geun Ho

AU - Song, Min Suk

AU - Chang, Jeesoo

AU - Min, Kyung Kyu

AU - Jang, Taejin

AU - Lee, Jong Ho

AU - Park, Byung Gook

AU - Kim, Hyungjin

PY - 2022/4/1

Y1 - 2022/4/1

N2 - In this work, we present a hardware neural network with capacitor-based synaptic devices. A capacitor-based synaptic device was developed using a MOS capacitor structure with a charge trapping layer. Due to the flat band voltage shift by charge trapping and its non-linear {C} - {V} characteristics, multilevel weight values could be implemented by the charge occurring when charging and discharging the capacitor. The vector-matrix multiplication (VMM) function was also experimentally verified using a fabricated synapse array based on NAND flash structure.

AB - In this work, we present a hardware neural network with capacitor-based synaptic devices. A capacitor-based synaptic device was developed using a MOS capacitor structure with a charge trapping layer. Due to the flat band voltage shift by charge trapping and its non-linear {C} - {V} characteristics, multilevel weight values could be implemented by the charge occurring when charging and discharging the capacitor. The vector-matrix multiplication (VMM) function was also experimentally verified using a fabricated synapse array based on NAND flash structure.

KW - Capacitor-based synaptic device

KW - MOS capacitor

KW - NAND flash memory

KW - capacitive neural network

KW - neuromorphic system

KW - spiking neural network (SNN)

UR - http://www.scopus.com/inward/record.url?scp=85124181706&partnerID=8YFLogxK

U2 - 10.1109/LED.2022.3149029

DO - 10.1109/LED.2022.3149029

M3 - Article

AN - SCOPUS:85124181706

SN - 0741-3106

VL - 43

SP - 549

EP - 552

JO - IEEE Electron Device Letters

JF - IEEE Electron Device Letters

IS - 4

ER -

Capacitor-Based Synaptic Devices for Hardware Spiking Neural Networks

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this