Paper 2020/239

Viktoria: A new Paradigm for Hash Functions

Edimar Veríssimo

Abstract

Viktoria hash is a compression function that generates a set of 512 bits from an arbitrary size input (limit of 2^480-1 bytes). This hash function contains some internal routines clearly inspired by AES and RC4 symmetric algorithms [14]. The new paradigm presents two major innovations: a fast preprocessing that initiates an internal state of 256!^2 permutations and a post-processing that guarantees a minimum number of executed rounds of 2^13. The pre-processing allows to differentiate very similar messages in the first runs of the algorithm. In the post-processing we have a safety barrier provided by a large number of rounds through a different structure of the main processing. The Viktoria algorithm seems to inaugurate a new design model in the construction of robust hash functions for some reasons, among them we highlight: the customization of the internal state according to each message, the elegance and efficiency of its main function and also a supposed high margin of safety provided by its post-processing function. Viktoria hash can also process bit oriented messages (whose last byte size is not complete) and generate larger hashes (1024, 1536, 2048 or larger) always as multiples of 512.

Note: Viktoria is a hash algorithm with an innovative design that seems to provide a high margin of safety.

Metadata
Available format(s)
PDF
Category
Applications
Publication info
Preprint.
Keywords
viktoriacompression functioncollisionhash functionirreversible functiondigital mapping of a message.
Contact author(s)
yugi386 @ yahoo com br
History
2020-02-25: received
Short URL
https://ia.cr/2020/239
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2020/239,
      author = {Edimar Veríssimo},
      title = {Viktoria: A new Paradigm for Hash Functions},
      howpublished = {Cryptology {ePrint} Archive, Paper 2020/239},
      year = {2020},
      url = {https://eprint.iacr.org/2020/239}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.