The technological advancements made in recent years regarding the transfer of personal data have brought along an imperative need for increased security and encryption capabilities. Information sent across these electronic platforms is most often intended to be delivered to and received by specific individuals. However, the notion that these selected individuals are the only people that are able to come into contact with the information is flawed. A more realistic assumption, and the assumption that is currently demonstrated, is that the information sent can and will be intercepted. This means that successful encryption of the data is an invaluable part of the transfer process. For this reason, the study of encryption and the validation of cryptographic functions are topics that computer scientists continually work to improve. Current practice for determining the success of cryptographic functions tends to consist of various statistical tests conducted on random outputs from the algorithm. In this thesis, we propose an array-based structure to validate not only the output from an cryptographic function, but the cryptographic function itself. In using an array-based structure such as this, we do not limit ourselves to only detecting output that suggests a failure for the encryption function. With this structure we allow an opportunity to detect specific contributors to the failure within the algorithm.

Library of Congress Subject Headings

Data encryption (Computer science); Statistical hypothesis testing

Publication Date


Document Type


Student Type


Degree Name

Applied Statistics (MS)

Department, Program, or Center

The John D. Hromi Center for Quality and Applied Statistics (KGCOE)


Peter Bajorski

Advisor/Committee Member

Alan Kaminsky

Advisor/Committee Member

Marcin Lukowiak


Physical copy available from RIT's Wallace Library at QA76.9.A25 H69 2016


RIT – Main Campus