A simulation study on the accuracy of cryptographic randomness tests

Demirhan, H and Bitirim, N 2017, 'A simulation study on the accuracy of cryptographic randomness tests', Simulation, vol. 93, no. 12, pp. 1113-1122.


Document type: Journal Article
Collection: Journal Articles

Title A simulation study on the accuracy of cryptographic randomness tests
Author(s) Demirhan, H
Bitirim, N
Year 2017
Journal name Simulation
Volume number 93
Issue number 12
Start page 1113
End page 1122
Total pages 10
Publisher Sage
Abstract Randomness provided by pseudo-random number generators is the one of the most vital parts of cryptographic applications. There are two gaps in the cryptographic randomness test procedures used to evaluate the degree of randomness. Firstly, although there are more accurate alternatives, the usual chi-square test is directly applied regardless of the predictive power of the tests. Secondly, although there are more than 100 cryptographic randomness tests available in the literature of cryptography, the statistical characteristics and accuracy of those hypothesis tests have not been figured out by an extensive simulation study. To conduct appropriate and reliable hypothesis tests, the main statistical characteristics of the tests should be studied. In this article, the usage of alternatives to the chi-square test, such as Anderson-Darling, Kolmogorov-Smirnov, and Jarque-Bera tests, in testing the cryptographic randomness is proposed to get better statistical properties. Also, the effects of type-I error, sensitivity, specificity, power, negative predictive value, and goodness-of-fit tests on the accuracy of recently proposed and existing cryptographic randomness tests are evaluated by an extensive Monte Carlo simulation study. The results are beneficial for practitioners wishing to choose the most appropriate cryptographic randomness test procedure and for the evaluation of accuracy of the cryptographic randomness tests in the detection of non-randomness for cryptographic applications.
Subject Applied Statistics
Statistical Theory
Keyword(s) Bit-length
hypothesis testing
negative predictive value
power
random number generator
sensitivity
significance level
specificity
type-I error
DOI - identifier 10.1177/0037549717726145
Copyright notice © The Author(s) 2017
ISSN 0037-5497
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in Thomson Reuters Web of Science Article
Scopus Citation Count Cited 0 times in Scopus Article
Altmetric details:
Access Statistics: 24 Abstract Views  -  Detailed Statistics
Created: Wed, 03 Jan 2018, 12:17:00 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us