Paper 2015/902

Differential Analysis on Simeck and SIMON with Dynamic Key-guessing Techniques

Kexin Qiao, Lei Hu, and Siwei Sun

Abstract

The Simeck family of lightweight block ciphers was proposed in CHES 2015 which combines the good design components from NSA designed ciphers SIMON and SPECK. Dynamic key-guessing techniques were proposed by Wang {\it et al.} to greatly reduce the key space guessed in differential cryptanalysis and work well on SIMON. In this paper, we implement the dynamic key-guessing techniques in a program to automatically give out the data in dynamic key-guessing procedure and thus simplify the security evaluation of SIMON and Simeck like block ciphers regarding differential attacks. We use the differentials from Kölbl {\it et al.}'s work and also a differential with lower Hamming weight we find using Mixed Integer Linear Programming method to attack 22-round Simeck32, 28-round Simeck48 and 35-round Simeck64. Besides, we launch the same attack procedure on four members of SIMON family by use of newly proposed differentials in CRYPTO2015 and get new attack results on 22-round SIMON32/64, 24-round SIMON48/96, 28, 29-round SIMON64/96 and 29, 30-round SIMON64/128. As far as we are concerned, our results on SIMON64 are currently the best results.

Note: We apply the method in the original paper to another cipher and add new results about this cipher.

Metadata
Available format(s)
PDF
Publication info
Preprint. MINOR revision.
Keywords
SimeckSIMONDynamic Key-guessingDifferential Cryptanalysis
Contact author(s)
qiaokexin @ iie ac cn
History
2016-04-14: revised
2015-09-17: received
See all versions
Short URL
https://ia.cr/2015/902
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2015/902,
      author = {Kexin Qiao and Lei Hu and Siwei Sun},
      title = {Differential Analysis on Simeck and {SIMON} with Dynamic Key-guessing Techniques},
      howpublished = {Cryptology {ePrint} Archive, Paper 2015/902},
      year = {2015},
      url = {https://eprint.iacr.org/2015/902}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.