Deep learning-based I-V global parameter extraction for BSIM-CMG

Fredo Chavez; Chien-Ting Tung; Ming-Yen Kao; Chenming Hu; Jen-Hao Chen; Sourabh Khandelwal

doi:10.1016/j.sse.2023.108766

Deep learning-based I-V global parameter extraction for BSIM-CMG

Fredo Chavez^*, Chien-Ting Tung, Ming-Yen Kao, Chenming Hu, Jen-Hao Chen, Sourabh Khandelwal

^*Corresponding author for this work

School of Engineering

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

11 Citations (Scopus)

374 Downloads (Pure)

Abstract

A new deep-learning-based parameter extraction for a global (multiple gate lengths) BSIM-CMG drain-current model is presented in this paper. The approach starts with generating 300K training dataset, consisting of 778 million data points to train the deep learning engine. Training data is generated by Monte Carlo simulation. The I-V data and the device geometry information from multiple devices serve to train a deep-learning (DL) model to predict BSIM-CMG parameters. The performance of DL-based extraction is verified by using the trained DL model to extract parameters of 10 nm FinFET technology simulated with TCAD. The DL-extracted BSIM-CMG model shows a good accuracy for eight different gate-lengths. The created BSIM-CMG global model was also able to reproduce the scalability in key electrical performance parameters such as off current I_off, saturation current I_sat, linear current I_lin and the threshold voltage in linear V_th,lin and saturation V_th,sat conditions. The developed solution significantly reduces the model extraction time for a global BSIM-CMG model. This new technique can expedite the development of process design kits (PDK).

Original language	English
Article number	108766
Pages (from-to)	1-5
Number of pages	5
Journal	Solid-State Electronics
Volume	209
DOIs	https://doi.org/10.1016/j.sse.2023.108766
Publication status	Published - Nov 2023

Bibliographical note

Copyright the Author(s) 2023. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

Parameter extraction
Berkeley Short-channel IGFET Model–Common Multi-Gate (BSIM-CMG)
Fin field-effect transistor (FinFET)
Deep learning

Access to Document

10.1016/j.sse.2023.108766Licence: CC BY

Publisher version (open access)Final published version, 2.3 MBLicence: CC BY

Cite this

@article{6de8b2200f8f464f851e9774ff7cd40b,

title = "Deep learning-based I-V global parameter extraction for BSIM-CMG",

abstract = "A new deep-learning-based parameter extraction for a global (multiple gate lengths) BSIM-CMG drain-current model is presented in this paper. The approach starts with generating 300K training dataset, consisting of 778 million data points to train the deep learning engine. Training data is generated by Monte Carlo simulation. The I-V data and the device geometry information from multiple devices serve to train a deep-learning (DL) model to predict BSIM-CMG parameters. The performance of DL-based extraction is verified by using the trained DL model to extract parameters of 10 nm FinFET technology simulated with TCAD. The DL-extracted BSIM-CMG model shows a good accuracy for eight different gate-lengths. The created BSIM-CMG global model was also able to reproduce the scalability in key electrical performance parameters such as off current Ioff, saturation current Isat, linear current Ilin and the threshold voltage in linear Vth,lin and saturation Vth,sat conditions. The developed solution significantly reduces the model extraction time for a global BSIM-CMG model. This new technique can expedite the development of process design kits (PDK).",

keywords = "Parameter extraction, Berkeley Short-channel IGFET Model–Common Multi-Gate (BSIM-CMG), Fin field-effect transistor (FinFET), Deep learning",

author = "Fredo Chavez and Chien-Ting Tung and Ming-Yen Kao and Chenming Hu and Jen-Hao Chen and Sourabh Khandelwal",

note = "Copyright the Author(s) 2023. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.",

year = "2023",

month = nov,

doi = "10.1016/j.sse.2023.108766",

language = "English",

volume = "209",

pages = "1--5",

journal = "Solid-State Electronics",

issn = "0038-1101",

publisher = "Elsevier",

}

TY - JOUR

T1 - Deep learning-based I-V global parameter extraction for BSIM-CMG

AU - Chavez, Fredo

AU - Tung, Chien-Ting

AU - Kao, Ming-Yen

AU - Hu, Chenming

AU - Chen, Jen-Hao

AU - Khandelwal, Sourabh

N1 - Copyright the Author(s) 2023. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2023/11

Y1 - 2023/11

N2 - A new deep-learning-based parameter extraction for a global (multiple gate lengths) BSIM-CMG drain-current model is presented in this paper. The approach starts with generating 300K training dataset, consisting of 778 million data points to train the deep learning engine. Training data is generated by Monte Carlo simulation. The I-V data and the device geometry information from multiple devices serve to train a deep-learning (DL) model to predict BSIM-CMG parameters. The performance of DL-based extraction is verified by using the trained DL model to extract parameters of 10 nm FinFET technology simulated with TCAD. The DL-extracted BSIM-CMG model shows a good accuracy for eight different gate-lengths. The created BSIM-CMG global model was also able to reproduce the scalability in key electrical performance parameters such as off current Ioff, saturation current Isat, linear current Ilin and the threshold voltage in linear Vth,lin and saturation Vth,sat conditions. The developed solution significantly reduces the model extraction time for a global BSIM-CMG model. This new technique can expedite the development of process design kits (PDK).

AB - A new deep-learning-based parameter extraction for a global (multiple gate lengths) BSIM-CMG drain-current model is presented in this paper. The approach starts with generating 300K training dataset, consisting of 778 million data points to train the deep learning engine. Training data is generated by Monte Carlo simulation. The I-V data and the device geometry information from multiple devices serve to train a deep-learning (DL) model to predict BSIM-CMG parameters. The performance of DL-based extraction is verified by using the trained DL model to extract parameters of 10 nm FinFET technology simulated with TCAD. The DL-extracted BSIM-CMG model shows a good accuracy for eight different gate-lengths. The created BSIM-CMG global model was also able to reproduce the scalability in key electrical performance parameters such as off current Ioff, saturation current Isat, linear current Ilin and the threshold voltage in linear Vth,lin and saturation Vth,sat conditions. The developed solution significantly reduces the model extraction time for a global BSIM-CMG model. This new technique can expedite the development of process design kits (PDK).

KW - Parameter extraction

KW - Berkeley Short-channel IGFET Model–Common Multi-Gate (BSIM-CMG)

KW - Fin field-effect transistor (FinFET)

KW - Deep learning

UR - http://www.scopus.com/inward/record.url?scp=85171325943&partnerID=8YFLogxK

U2 - 10.1016/j.sse.2023.108766

DO - 10.1016/j.sse.2023.108766

M3 - Article

AN - SCOPUS:85171325943

SN - 0038-1101

VL - 209

SP - 1

EP - 5

JO - Solid-State Electronics

JF - Solid-State Electronics

M1 - 108766

ER -

Deep learning-based I-V global parameter extraction for BSIM-CMG

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this