Sub-nA Low-Current HZO Ferroelectric Tunnel Junction for High-Performance and Accurate Deep Learning Acceleration

Tzu Yun Wu, Tian-Sheuan Chang, Heng Yuan Lee, Shyh Shyuan Sheu, Wei Chung Lo, Tuo-Hung Hou*, Hsin Hui Huang, Yueh Hua Chu, Chih Cheng Chang, Ming Hung Wu, Chien Hua Hsu, Chien Ting Wu, Min Ci Wu, Wen-Wei Wu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

This paper presents a unique opportunity of HZO ferroelectric tunnel junction (FTJ) for in-memory computing. The device operates at an extremely low sub-nA current while simultaneously achieving 50-ns fast switching, > 107 cycling endurance, > 10-yr retention, minimal variability, and analog state modulation. We analyze an FTJ-based deep binary neural network. It achieves better accuracy and remarkable 702, 101, and 7×104 times improvements in power, area, and energy-area product efficiency compared with those using NVMs with a typical μA cell current designed for fast memory access.

Original languageEnglish
Title of host publication2019 IEEE International Electron Devices Meeting, IEDM 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages4
ISBN (Electronic)9781728140315
DOIs
StatePublished - Dec 2019
Event65th Annual IEEE International Electron Devices Meeting, IEDM 2019 - San Francisco, United States
Duration: 7 Dec 201911 Dec 2019

Publication series

NameTechnical Digest - International Electron Devices Meeting, IEDM
Volume2019-December
ISSN (Print)0163-1918

Conference

Conference65th Annual IEEE International Electron Devices Meeting, IEDM 2019
CountryUnited States
CitySan Francisco
Period7/12/1911/12/19

Fingerprint Dive into the research topics of 'Sub-nA Low-Current HZO Ferroelectric Tunnel Junction for High-Performance and Accurate Deep Learning Acceleration'. Together they form a unique fingerprint.

Cite this