G-PUF: An Intrinsic PUF Based on GPU Error Signatures

Bruno Forlin; Ronaldo  Husemann; Luigi Carro; Cezar Reinbrecht; Said Hamdioui; Mottaqiallah Taouil

doi:10.1109/ETS48528.2020.9131562

G-PUF: An Intrinsic PUF Based on GPU Error Signatures

Bruno Forlin, Ronaldo Husemann, Luigi Carro, Cezar Reinbrecht, Said Hamdioui, Mottaqiallah Taouil

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

2 Citations (Scopus)

Abstract

Physically Unclonable Functions (PUFs) are security primitives that provide trustworthy hardware for key-generation and device authentication. Among them, in contrast to dedicated PUFs, intrinsic PUFs are created from existing hardware components that exploit their variability through software. In this work we focus on GPUs and present G-PUF, a PUF implemented entirely in software on CUDA and hence does not require hardware modifications. Our results show that G-PUF has comparable characteristics to SRAM and DRAM PUFs in terms of uniformity 55.61% and reliability 90.09%.

Original language	English
Title of host publication	2020 IEEE European Test Symposium (ETS)
Subtitle of host publication	Proceedings
Publisher	IEEE
Pages	1-2
Number of pages	2
ISBN (Electronic)	978-1-7281-4312-5
ISBN (Print)	978-1-7281-4313-2
DOIs	https://doi.org/10.1109/ETS48528.2020.9131562
Publication status	Published - 2020
Event	ETS 2020: 2020 IEEE European Test Symposium - Tallinn, Estonia Duration: 25 May 2020 → 29 May 2020

Conference

Conference	ETS 2020
Country/Territory	Estonia
City	Tallinn
Period	25/05/20 → 29/05/20

Keywords

G-PUF
GPU
Intrinsic PUF
Physically Unclonable Functions
Security

Access to Document

10.1109/ETS48528.2020.9131562

Cite this

@inproceedings{18234f6f745a4433aa628b3ffba29696,

title = "G-PUF: An Intrinsic PUF Based on GPU Error Signatures",

abstract = "Physically Unclonable Functions (PUFs) are security primitives that provide trustworthy hardware for key-generation and device authentication. Among them, in contrast to dedicated PUFs, intrinsic PUFs are created from existing hardware components that exploit their variability through software. In this work we focus on GPUs and present G-PUF, a PUF implemented entirely in software on CUDA and hence does not require hardware modifications. Our results show that G-PUF has comparable characteristics to SRAM and DRAM PUFs in terms of uniformity 55.61% and reliability 90.09%.",

keywords = "G-PUF, GPU, Intrinsic PUF, Physically Unclonable Functions, Security",

author = "Bruno Forlin and Ronaldo Husemann and Luigi Carro and Cezar Reinbrecht and Said Hamdioui and Mottaqiallah Taouil",

year = "2020",

doi = "10.1109/ETS48528.2020.9131562",

language = "English",

isbn = "978-1-7281-4313-2",

pages = "1--2",

booktitle = "2020 IEEE European Test Symposium (ETS)",

publisher = "IEEE",

address = "United States",

note = "ETS 2020 : 2020 IEEE European Test Symposium ; Conference date: 25-05-2020 Through 29-05-2020",

}

TY - GEN

T1 - G-PUF

T2 - ETS 2020

AU - Forlin, Bruno

AU - Husemann, Ronaldo

AU - Carro, Luigi

AU - Reinbrecht, Cezar

AU - Hamdioui, Said

AU - Taouil, Mottaqiallah

PY - 2020

Y1 - 2020

N2 - Physically Unclonable Functions (PUFs) are security primitives that provide trustworthy hardware for key-generation and device authentication. Among them, in contrast to dedicated PUFs, intrinsic PUFs are created from existing hardware components that exploit their variability through software. In this work we focus on GPUs and present G-PUF, a PUF implemented entirely in software on CUDA and hence does not require hardware modifications. Our results show that G-PUF has comparable characteristics to SRAM and DRAM PUFs in terms of uniformity 55.61% and reliability 90.09%.

AB - Physically Unclonable Functions (PUFs) are security primitives that provide trustworthy hardware for key-generation and device authentication. Among them, in contrast to dedicated PUFs, intrinsic PUFs are created from existing hardware components that exploit their variability through software. In this work we focus on GPUs and present G-PUF, a PUF implemented entirely in software on CUDA and hence does not require hardware modifications. Our results show that G-PUF has comparable characteristics to SRAM and DRAM PUFs in terms of uniformity 55.61% and reliability 90.09%.

KW - G-PUF

KW - GPU

KW - Intrinsic PUF

KW - Physically Unclonable Functions

KW - Security

UR - http://www.scopus.com/inward/record.url?scp=85089145792&partnerID=8YFLogxK

U2 - 10.1109/ETS48528.2020.9131562

DO - 10.1109/ETS48528.2020.9131562

M3 - Conference contribution

AN - SCOPUS:85089145792

SN - 978-1-7281-4313-2

SP - 1

EP - 2

BT - 2020 IEEE European Test Symposium (ETS)

PB - IEEE

Y2 - 25 May 2020 through 29 May 2020

ER -

G-PUF: An Intrinsic PUF Based on GPU Error Signatures

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this