A Non-linear Input Converter Inversely Pre-distorted Against Nonlinear Behavior of FG-based Neuromorphic Synaptic Devices
- Authors
- Hwang, J.-Y.[Hwang, J.-Y.]; Kwon, K.-W.[Kwon, K.-W.]
- Issue Date
- 2021
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- CT memory; Floating gate; non-Linearity
- Citation
- Proceedings - International SoC Design Conference 2021, ISOCC 2021, pp.409 - 410
- Indexed
- SCOPUS
- Journal Title
- Proceedings - International SoC Design Conference 2021, ISOCC 2021
- Start Page
- 409
- End Page
- 410
- URI
- https://scholarworks.bwise.kr/skku/handle/2021.sw.skku/95425
- DOI
- 10.1109/ISOCC53507.2021.9614035
- ISSN
- 0000-0000
- Abstract
- In this paper, we propose a non-linear input converter inversely pre-distorted. In terms of the cell of FG-based neuromorphic synaptic devices, MOSFET read current of nonlinearity problem is caused by two individual inputs(FG weight input voltage). Against non-linear convex output current, concave I-V characteristics of common source amp attributed to its high order terms in low input voltage region improves nonlinearity. To show how much linearizing the nonlinear convex curve of read current, we compare coefficient of determination(R2) of conventional cell, additional cap between FG and Drain of MOSFET in the cell and proposed distorted input amp to cell individually. Briefly, linearity factor(R2) of this work shows 0.99 and this value closes to 1 than other cases. © 2021 IEEE.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - Information and Communication Engineering > Department of Semiconductor Systems Engineering > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/skku/handle/2021.sw.skku/95425)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.