User: Guest  Login
Original title:
Metrics for physical unclonable functions
Translated title:
Metriken für Physical Unclonable Functions
Author:
Wilde, Florian K. A.
Year:
2021
Document type:
Dissertation
Faculty/School:
Fakultät für Elektrotechnik und Informationstechnik
Advisor:
Sigl, Georg (Prof. Dr.)
Referee:
Sigl, Georg (Prof. Dr.); Mukhopadhyay, Debdeep (Prof., Ph.D.)
Language:
en
Subject group:
DAT Datenverarbeitung, Informatik; ELT Elektrotechnik
Keywords:
physical unclonable function, PUF, response mass function, RMF, expected conditional min-entropy, challengeability class, spatial autocorrelation, spatial correlation, SPACA, hypothesis test, metric, statistics, multivariate, entropy, min-entropy, mutual information, conditional entropy, helper data system, HDS, helper data algorithm, HDA, dataset, real world, histogram, Hamming distance, Hamming weight, compression, confidence interval, principal component analysis, PCA
TUM classification:
DAT 460
Abstract:
Uni- and multivariate statistical models are used to compare a variety of existing metrics for the security evaluation of physical unclonable function (PUF) candidates, e.g. for expectation, variance, confidence intervals. As methodical improvement, statistical hypothesis testing with PUF specific tests is proposed, e.g. for bias or spatial autocorrelation. The response mass function (RMF) is introduced to represent the probability distribution of PUF responses even with thousands of bits.
Translated abstract:
Mittels uni- und multivariater statistischer Modelle wird eine Auswahl bestehender Metriken zur Sicherheitsanalyse von Physical Unclonable Functions (PUFs) verglichen, z.B. bezüglich Erwartungswert, Varianz und Konfidenzintervallen. Als methodische Verbesserung werden statistische Hypothesentests mit PUF spezifischen Tests vorgeschlagen, z.B. auf Bias oder räumliche Autokorrelation. Die Antworthäufigkeitsfunktion wird eingeführt, um die Wahrscheinlichkeitsverteilung der PUF Antwort auch bei mehr...     »
WWW:
https://mediatum.ub.tum.de/?id=1612868
Date of submission:
10.06.2021
Oral examination:
03.12.2021
File size:
4564870 bytes
Pages:
256
Urn (citeable URL):
https://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:91-diss-20211203-1612868-1-6
Last change:
17.02.2022
 BibTeX