Logo image
28 nm HKMG-Based Current Limited FeFET Crossbar-Array for Inference Application
期刊文章   開放取用(OA)   同儕審查

28 nm HKMG-Based Current Limited FeFET Crossbar-Array for Inference Application

Sourav De, Franz Muller, Sunanda Thunder, Sukhrob Abdulazhanov, Nellie Laleni, Maximilian Lederer, Tarek Ali, Yannick Raffel, Stefan Dunkel, Shaown Mojumder, …
IEEE Transactions on Electron Devices, 卷.69(12), 頁碼.7194-7198
12/2022

摘要

Ferroelectric field-effect transistor (FeFET) ferroelectric memory hafnium oxide (HfO-) inference neuromorphic Electronic Optical and Magnetic Materials Electrical and Electronic Engineering
This article reports a novel ferroelectric field-effect transistor (FeFET)-based crossbar array cascaded with an external resistor. The external resistor is shunted with the column of the FeFET array, as a current limiter and reduces the impact of variations in drain current (Id), especially in a low threshold voltage (LVT) state. We have designed crossbar arrays of 8×8 sizes and performed multiply-and-accumulate (MAC) operations. Furthermore, we have evaluated the performance of the current limited FeFET crossbar array in system-level applications. Finally, the system-level performance evaluation was done by neuromorphic simulation of the resistor-shunted FeFET crossbar array. The crossbar array achieved software-comparable inference accuracy (97%) for National Institute of Standards and Technology (MNIST) datasets with multilayer perceptron (MLP) neural network, whereas the crossbar arrays built solely with FeFETs failed to learn, yielding only 9.8% accuracy.

檔案與連結 (1)

url
https://doi.org/10.1109/TED.2022.3216973檢視
已出版(紀錄版本) 開放

相關連結

指標

1 檢視次數

詳細資料

Logo image