Paper Info
Title
Delay PUF Assessment Method Based on Side-Channel and Modeling Analyzes: The Final Piece of All-in-One Assessment Methodology
Abstract
Most of the recent literature investigates mathematical analyses performed on the simulated behavior of the delay physically unclonable function (PUF). These analyses are not sufficient to characterize the PUF leakage under actual operating conditions, and it is hard to model the distribution of the noise that might confuse the assessment of the PUF. Furthermore, simulated analyses may lead to an overestimation of the risk. We show that the availability of an embedded image of the target PUF is important for assessing security against real attacks by considering different input conditions. In this paper, we propose a delay PUF assessment method that combines side-channel analyses and modeling analyses. Our proposed method is the last piece of the all-in-one methodology to assess the security of the delay PUF. Finally, we conduct an experiment to evaluate a real arbiter-PUF (APUF) with our proposed assessment method. Our result shows that a modeling analysis using random forest (RF) and advanced support vector machine (SVM) are suitable for the APUF in the assessment method.
Year
DOI
Venue
2016
10.1109/TrustCom.2016.0064
2016 IEEE Trustcom/BigDataSE/ISPA
Keywords
Field
DocType
physically unclonable function (PUF),PUF Assessment,side-channel analysis,modeling analysis
Data mining,Computer science,Computer security,Support vector machine,Robustness (computer science),Side channel attack,Physical unclonable function,Random forest
Conference
ISBN
Citations 
PageRank 
978-1-5090-3206-8
0
0.34
References 
Authors
11
9
Name
Order
Citations
PageRank
Kazuhide Fukushima117829.48
Youssef Souissi200.34
Seira Hidano300.68
Robert Nguyen400.34
Jean-Luc Danger500.34
Sylvain Guilley6279.46
Yuto Nakano7206.65
Shinsaku Kiyomoto833264.29
Laurent Sauvage926217.81