On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs

Ghazanfar  Ali; Leila  Bagheriye; Hans Manhaeve; Hans Gerard Kerkhoff

doi:10.1109/ATS49688.2020.9301509

On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs

Ghazanfar Ali, Leila Bagheriye, Hans Manhaeve, Hans Gerard Kerkhoff

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

5 Downloads (Pure)

Abstract

The usage of embedded instruments (EIs) in a processor core to address dependability challenges of modern-day Multi-Processor System-on-Chip (MP-SoC) has been studied in literature extensively. Data from these EIs can be used in applications like end-of-lifetime (EOL) predictions. However, inaccuracies present in the data from these EIs, due to their selfaging and resolution limitations during digitization, can lead to an inaccurate EOL assessment. In this paper, it is presented that in the presence of such inaccuracies from EIs as well as correlation between EIs, principal component analysis (PCA) based datafusion approach for determining the EOL of selected critical paths provided overall better EOL predictions as compared to EOL predictions based on standalone EIs. Verification was performed with a commercial software-based EOL predictor tool ARULE running on a personal computer. Moreover, the presented results on the computational requirements for the presented data-fusion approach showed little overhead in terms of memory, execution time and energy requirements.

Original language	English
Title of host publication	IEEE Asian Test Symposium, ATS 2020
Place of Publication	Piscataway, NJ
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	978-1-7281-7467-9
ISBN (Print)	978-1-7281-7468-6
DOIs	https://doi.org/10.1109/ATS49688.2020.9301509
Publication status	Published - 23 Nov 2020
Event	29th IEEE Asian Test Symposium, ATS 2020 - Penang, Malaysia Duration: 23 Nov 2020 → 26 Nov 2020 Conference number: 29

Publication series

Name	IEEE Asian Test Symposium (ATS)
Publisher	IEEE
Number	29
Volume	2020
ISSN (Print)	1081-7735
ISSN (Electronic)	2377-5386

Conference

Conference	29th IEEE Asian Test Symposium, ATS 2020
Abbreviated title	ATS
Country/Territory	Malaysia
City	Penang
Period	23/11/20 → 26/11/20

Keywords

Dependability
Embedded instruments
Data fusion
Life-time prognostics
Machine learning
n/a OA procedure

Access to Document

10.1109/ATS49688.2020.9301509

Cite this

@inproceedings{e29a953c04204b7da38f5f7880693928,

title = "On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs",

abstract = "The usage of embedded instruments (EIs) in a processor core to address dependability challenges of modern-day Multi-Processor System-on-Chip (MP-SoC) has been studied in literature extensively. Data from these EIs can be used in applications like end-of-lifetime (EOL) predictions. However, inaccuracies present in the data from these EIs, due to their selfaging and resolution limitations during digitization, can lead to an inaccurate EOL assessment. In this paper, it is presented that in the presence of such inaccuracies from EIs as well as correlation between EIs, principal component analysis (PCA) based datafusion approach for determining the EOL of selected critical paths provided overall better EOL predictions as compared to EOL predictions based on standalone EIs. Verification was performed with a commercial software-based EOL predictor tool ARULE running on a personal computer. Moreover, the presented results on the computational requirements for the presented data-fusion approach showed little overhead in terms of memory, execution time and energy requirements.",

keywords = "Dependability, Embedded instruments, Data fusion, Life-time prognostics, Machine learning, n/a OA procedure",

author = "Ghazanfar Ali and Leila Bagheriye and Hans Manhaeve and Kerkhoff, {Hans Gerard}",

year = "2020",

month = nov,

day = "23",

doi = "10.1109/ATS49688.2020.9301509",

language = "English",

isbn = "978-1-7281-7468-6",

series = "IEEE Asian Test Symposium (ATS)",

publisher = "IEEE",

number = "29",

booktitle = "IEEE Asian Test Symposium, ATS 2020",

address = "United States",

note = "29th IEEE Asian Test Symposium, ATS 2020, ATS ; Conference date: 23-11-2020 Through 26-11-2020",

}

Ali, G, Bagheriye, L, Manhaeve, H & Kerkhoff, HG 2020, On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs. in IEEE Asian Test Symposium, ATS 2020. IEEE Asian Test Symposium (ATS), no. 29, vol. 2020, IEEE, Piscataway, NJ, 29th IEEE Asian Test Symposium, ATS 2020, Penang, Malaysia, 23/11/20. https://doi.org/10.1109/ATS49688.2020.9301509

On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs. / Ali, Ghazanfar ; Bagheriye, Leila ; Manhaeve, Hans et al.
IEEE Asian Test Symposium, ATS 2020. Piscataway, NJ: IEEE, 2020. (IEEE Asian Test Symposium (ATS); Vol. 2020, No. 29).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs

AU - Ali, Ghazanfar

AU - Bagheriye, Leila

AU - Manhaeve, Hans

AU - Kerkhoff, Hans Gerard

N1 - Conference code: 29

PY - 2020/11/23

Y1 - 2020/11/23

N2 - The usage of embedded instruments (EIs) in a processor core to address dependability challenges of modern-day Multi-Processor System-on-Chip (MP-SoC) has been studied in literature extensively. Data from these EIs can be used in applications like end-of-lifetime (EOL) predictions. However, inaccuracies present in the data from these EIs, due to their selfaging and resolution limitations during digitization, can lead to an inaccurate EOL assessment. In this paper, it is presented that in the presence of such inaccuracies from EIs as well as correlation between EIs, principal component analysis (PCA) based datafusion approach for determining the EOL of selected critical paths provided overall better EOL predictions as compared to EOL predictions based on standalone EIs. Verification was performed with a commercial software-based EOL predictor tool ARULE running on a personal computer. Moreover, the presented results on the computational requirements for the presented data-fusion approach showed little overhead in terms of memory, execution time and energy requirements.

AB - The usage of embedded instruments (EIs) in a processor core to address dependability challenges of modern-day Multi-Processor System-on-Chip (MP-SoC) has been studied in literature extensively. Data from these EIs can be used in applications like end-of-lifetime (EOL) predictions. However, inaccuracies present in the data from these EIs, due to their selfaging and resolution limitations during digitization, can lead to an inaccurate EOL assessment. In this paper, it is presented that in the presence of such inaccuracies from EIs as well as correlation between EIs, principal component analysis (PCA) based datafusion approach for determining the EOL of selected critical paths provided overall better EOL predictions as compared to EOL predictions based on standalone EIs. Verification was performed with a commercial software-based EOL predictor tool ARULE running on a personal computer. Moreover, the presented results on the computational requirements for the presented data-fusion approach showed little overhead in terms of memory, execution time and energy requirements.

KW - Dependability

KW - Embedded instruments

KW - Data fusion

KW - Life-time prognostics

KW - Machine learning

KW - n/a OA procedure

U2 - 10.1109/ATS49688.2020.9301509

DO - 10.1109/ATS49688.2020.9301509

M3 - Conference contribution

SN - 978-1-7281-7468-6

T3 - IEEE Asian Test Symposium (ATS)

BT - IEEE Asian Test Symposium, ATS 2020

PB - IEEE

CY - Piscataway, NJ

T2 - 29th IEEE Asian Test Symposium, ATS 2020

Y2 - 23 November 2020 through 26 November 2020

ER -

On-chip EOL Prognostics Using Data-Fusion of Embedded Instruments for Dependable MP-SoCs

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this