Multiple sensitive volume based soft error rate estimation with machine learning

Soichi Hirokawa, Ryo Harada, Kenshiro Sakuta, Yukinobu Watanabe, Masanori Hashimoto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

We propose a new methodology for soft error rate estimation using multiple sensitive volumes and machine learning. The proposed methodology assigns multiple sensitive volumes to a unit circuit (e.g. SRAM cell) and constructs a discriminator from TCAD simulations by machine learning. For each ion reproduced by radiation transport simulation, the discriminator judges whether an upset occurs or not, and consequently we can obtain soft error rate by counting the number of events judged as upset events. Advantages of the proposed methodology are: (1) empirical construction and adjustment of sensitive volume and critical charge is no longer necessary, (2) multiple transistors can be easily considered, and (3) event-wise accuracy can be improved. We confirmed the correlation between irradiation results and simulation results for 65-nm silicon on thin buried oxide (SOTB) SRAM. The estimation error was 7% without any empirical optimization of sensitive volume and critical charge.

Original languageEnglish
Title of host publication2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-4
Number of pages4
Volume2016-September
ISBN (Electronic)9781509043668
DOIs
Publication statusPublished - Oct 31 2017
Event16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016 - Bremen, Germany
Duration: Sep 19 2016Sep 23 2016

Other

Other16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016
CountryGermany
CityBremen
Period9/19/169/23/16

Fingerprint

machine learning
Discriminators
Static random access storage
Learning systems
discriminators
methodology
Error analysis
Transistors
Irradiation
Radiation
Silicon
Oxides
radiation transport
simulation
Networks (circuits)
Ions
learning
counting
transistors
adjusting

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Radiation

Cite this

Hirokawa, S., Harada, R., Sakuta, K., Watanabe, Y., & Hashimoto, M. (2017). Multiple sensitive volume based soft error rate estimation with machine learning. In 2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016 (Vol. 2016-September, pp. 1-4). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RADECS.2016.8093181

Multiple sensitive volume based soft error rate estimation with machine learning. / Hirokawa, Soichi; Harada, Ryo; Sakuta, Kenshiro; Watanabe, Yukinobu; Hashimoto, Masanori.

2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016. Vol. 2016-September Institute of Electrical and Electronics Engineers Inc., 2017. p. 1-4.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hirokawa, S, Harada, R, Sakuta, K, Watanabe, Y & Hashimoto, M 2017, Multiple sensitive volume based soft error rate estimation with machine learning. in 2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016. vol. 2016-September, Institute of Electrical and Electronics Engineers Inc., pp. 1-4, 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016, Bremen, Germany, 9/19/16. https://doi.org/10.1109/RADECS.2016.8093181
Hirokawa S, Harada R, Sakuta K, Watanabe Y, Hashimoto M. Multiple sensitive volume based soft error rate estimation with machine learning. In 2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016. Vol. 2016-September. Institute of Electrical and Electronics Engineers Inc. 2017. p. 1-4 https://doi.org/10.1109/RADECS.2016.8093181
Hirokawa, Soichi ; Harada, Ryo ; Sakuta, Kenshiro ; Watanabe, Yukinobu ; Hashimoto, Masanori. / Multiple sensitive volume based soft error rate estimation with machine learning. 2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016. Vol. 2016-September Institute of Electrical and Electronics Engineers Inc., 2017. pp. 1-4
@inproceedings{59fe89c67bea43dbbaa9bbf706c8c21f,
title = "Multiple sensitive volume based soft error rate estimation with machine learning",
abstract = "We propose a new methodology for soft error rate estimation using multiple sensitive volumes and machine learning. The proposed methodology assigns multiple sensitive volumes to a unit circuit (e.g. SRAM cell) and constructs a discriminator from TCAD simulations by machine learning. For each ion reproduced by radiation transport simulation, the discriminator judges whether an upset occurs or not, and consequently we can obtain soft error rate by counting the number of events judged as upset events. Advantages of the proposed methodology are: (1) empirical construction and adjustment of sensitive volume and critical charge is no longer necessary, (2) multiple transistors can be easily considered, and (3) event-wise accuracy can be improved. We confirmed the correlation between irradiation results and simulation results for 65-nm silicon on thin buried oxide (SOTB) SRAM. The estimation error was 7{\%} without any empirical optimization of sensitive volume and critical charge.",
author = "Soichi Hirokawa and Ryo Harada and Kenshiro Sakuta and Yukinobu Watanabe and Masanori Hashimoto",
year = "2017",
month = "10",
day = "31",
doi = "10.1109/RADECS.2016.8093181",
language = "English",
volume = "2016-September",
pages = "1--4",
booktitle = "2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Multiple sensitive volume based soft error rate estimation with machine learning

AU - Hirokawa, Soichi

AU - Harada, Ryo

AU - Sakuta, Kenshiro

AU - Watanabe, Yukinobu

AU - Hashimoto, Masanori

PY - 2017/10/31

Y1 - 2017/10/31

N2 - We propose a new methodology for soft error rate estimation using multiple sensitive volumes and machine learning. The proposed methodology assigns multiple sensitive volumes to a unit circuit (e.g. SRAM cell) and constructs a discriminator from TCAD simulations by machine learning. For each ion reproduced by radiation transport simulation, the discriminator judges whether an upset occurs or not, and consequently we can obtain soft error rate by counting the number of events judged as upset events. Advantages of the proposed methodology are: (1) empirical construction and adjustment of sensitive volume and critical charge is no longer necessary, (2) multiple transistors can be easily considered, and (3) event-wise accuracy can be improved. We confirmed the correlation between irradiation results and simulation results for 65-nm silicon on thin buried oxide (SOTB) SRAM. The estimation error was 7% without any empirical optimization of sensitive volume and critical charge.

AB - We propose a new methodology for soft error rate estimation using multiple sensitive volumes and machine learning. The proposed methodology assigns multiple sensitive volumes to a unit circuit (e.g. SRAM cell) and constructs a discriminator from TCAD simulations by machine learning. For each ion reproduced by radiation transport simulation, the discriminator judges whether an upset occurs or not, and consequently we can obtain soft error rate by counting the number of events judged as upset events. Advantages of the proposed methodology are: (1) empirical construction and adjustment of sensitive volume and critical charge is no longer necessary, (2) multiple transistors can be easily considered, and (3) event-wise accuracy can be improved. We confirmed the correlation between irradiation results and simulation results for 65-nm silicon on thin buried oxide (SOTB) SRAM. The estimation error was 7% without any empirical optimization of sensitive volume and critical charge.

UR - http://www.scopus.com/inward/record.url?scp=85043587481&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043587481&partnerID=8YFLogxK

U2 - 10.1109/RADECS.2016.8093181

DO - 10.1109/RADECS.2016.8093181

M3 - Conference contribution

AN - SCOPUS:85043587481

VL - 2016-September

SP - 1

EP - 4

BT - 2016 16th European Conference on Radiation and Its Effects on Components and Systems, RADECS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -