Counting the number of active S-boxes is a common way to evaluate the security of symmetric key cryptographic schemes against differential attack. Based on Mixed Integer Linear Programming (MILP), Mouha et al. proposed a method to accomplish this task automatically for word-oriented symmetric-key ciphers with SPN structures. However, this method can not be applied directly to block ciphers of SPN structures with bitwise permutation diffusion layers (S-bP structures), due to its ignorance of the diffusion effect derived collaboratively by nonlinear substitution layers and bitwise permutation layers. In this paper we extend Mouha et al.’s method for S-bP structures by introducing new representations for exclusive-or (XOR) differences to describe bit/word level differences simultaneously and by taking the collaborative diffusion effect of S-boxes and bitwise permutations into account. Our method is applied to the block cipher PRESENT-80, an international standard for lightweight symmetric key cryptography, to automatically evaluate its security against differential attacks. We obtain lower bounds on the numbers of active S-boxes in the single-key model for full 31-round PRESENT-80 and in related-key model for round-reduced PRESENT-80 up to 12 rounds, and therefore automatically prove that the full-round PRESENT-80 is secure against single-key differential attack, and the cost of related-key differential attack on the full-round PRESENT-80 is close to that of an exhaustive search: the best related-key differential characteristic for full PRESENT-80 is upper bounded by 2-72.
CITATION STYLE
Sun, S., Hu, L., Song, L., Xie, Y., & Wang, P. (2014). Automatic security evaluation of block ciphers with S-bP structures against related-key differential attacks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8567, pp. 39–51). Springer Verlag. https://doi.org/10.1007/978-3-319-12087-4_3
Mendeley helps you to discover research relevant for your work.