Please use this identifier to cite or link to this item:
Title: Implementation and testing of hash function based on modified skein algorithm
Authors: Barybin, Oleksii
Tkachenko, Vira
Zhabska, Yelyzaveta
Petrova, Iryna
Keywords: Hash functions
Statistical methods of analysis of hash functions
Issue Date: 2018
Abstract: Context and objective. The article shows that the original Skein hash algorithm, which was developed as part of the competition for the new standard SHA-3 of the National Institute of Standards and Technology USA (NIST) and was one of five finalists can be modified to improve its operation efficiency in systems with multicore processors, which are used in almost all modern desktop and mobile devices. Analysis of the few publications that dedicated to such improvement of the original algorithm shows that all the authors focus on improving efficiency by optimizing the algorithm computations parallelization because computing speed was one of its main weaknesses. The simplification the original algorithm was not considered before and became the main purpose of the article. Research methods. Analysis of the original algorithm showed that the basic premise for the modifications is flexibility for adjusting to build a hash function, particularly hash function is based on block encryption algorithm with adjustable block size. Block size and key are fixed, but they do not determine the size of the original hash function string, which complicates the method of attack, based on finding the key length. Alternating nature of the output line allows you to make hash function very flexible: it is possible to use a hash function and applications in devices with limited memory. Including mobile devices. Results. Modification for Skein based on the fact that the original algorithm is assumed that the hash process can be represented as a multilevel tree structure. The basic idea is that the message on the first hashing level is divided into blocks, for each of the blocks calculated parameters for the second and higher levels become the input values for the next level. A significant simplification has been proposed no speed up the algorithm – use the only single-level hash that actually means abandoning the use of tree structure calculations. Hash function based on the modified algorithm is implemented with the use of the programming language Python. Test results according to NIST SP 800-22 standard for set of 1,000 files with hash function results showed the percentage of files that have been tested successfully from 98.0 to 99.6% that indicates that the hash function under consideration meets random and pseudorandom generator integers and hash functions conditions. The scientific novelty and practical significance. It is the first time then simplification of the structure for Skein hash algorithm calculations effectiveness improvement was proposed and it was shown that the hash function based on a simplified algorithm comply NIST SP 800-22.
Appears in Collections:Бібліографічні матеріали

Files in This Item:
File Description SizeFormat 
4_PDF_Проблеми інформаційних технологій.pdf184,39 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.