Paper 2023/443

Abstraction Model of Probing and DFA Attacks on Block Ciphers

Yuiko Matsubara, University of Electro-Communications
Daiki Miyahara, University of Electro-Communications, National Institute of Advanced Industrial Science and Technology
Yohei Watanabe, University of Electro-Communications, National Institute of Advanced Industrial Science and Technology
Mitsugu Iwamoto, University of Electro-Communications
Kazuo Sakiyama, University of Electro-Communications
Abstract

A thread of physical attacks that try to obtain secret information from cryptographic modules has been of academic and practical interest. One of the concerns is determining its efficiency, e.g., the number of attack trials to recover the secret key. However, the accurate estimation of the attack efficiency is generally expensive because of the complexity of the physical attack on a cryptographic algorithm. Based on this background, in this study, we propose a new abstraction model for evaluating the attack efficiency of the probing and DFA attacks. The proposed model includes an abstracted attack target and attacker to determine the amount of leaked information obtained in a single attack trial. We can adapt the model flexibly to various attack scenarios and can get the attack efficiency quickly and precisely. In the probing attack on AES, the difference in the attack efficiency is only approximately 0.3% between the model and experimental values, whereas that of a previous model is approximately 16%. We also apply the probing attack on DES, and the results show that DES has a high resistance to the probing attack. Moreover, the proposed model works accurately also for the DFA attack on AES.

Metadata
Available format(s)
PDF
Category
Implementation
Publication info
Preprint.
Keywords
Physical attackProbing attackDifferential fault analysisAdvanced encryption standardInformation leakage
Contact author(s)
miyahara @ uec ac jp
History
2023-03-27: approved
2023-03-27: received
See all versions
Short URL
https://ia.cr/2023/443
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2023/443,
      author = {Yuiko Matsubara and Daiki Miyahara and Yohei Watanabe and Mitsugu Iwamoto and Kazuo Sakiyama},
      title = {Abstraction Model of Probing and DFA Attacks on Block Ciphers},
      howpublished = {Cryptology ePrint Archive, Paper 2023/443},
      year = {2023},
      note = {\url{https://eprint.iacr.org/2023/443}},
      url = {https://eprint.iacr.org/2023/443}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.