Abstract: At the Annual International Cryptology Conference in 2019, Gohr introduced a deep learning based cryptanalysis technique applicable to the reduced-round lightweight block ciphers with a short block of SPECK32/64. One significant challenge left unstudied by Gohr’s work is the implementation of key recovery attacks on large-state block ciphers based on deep learning. The purpose of this paper is to present an improved deep learning based framework for recovering keys for large-state block ciphers. First, we propose a key bit sensitivity test (KBST) based on deep learning to divide the key space objectively. Second, we propose a new method for constructing neural distinguisher combinations to improve a deep learning based key recovery framework for large-state block ciphers and demonstrate its rationality and effectiveness from the perspective of cryptanalysis. Under the improved key recovery framework, we train an efficient neural distinguisher combination for each large-state member of SIMON and SPECK and finally carry out a practical key recovery attack on the large-state members of SIMON and SPECK. Furthermore, we propose that the 13-round SIMON64 attack is the most effective approach for practical key recovery to date. Noteworthly, this is the first attempt to propose deep learning based practical key recovery attacks on 18-round SIMON128, 19-round SIMON128, 14-round SIMON96, and 14-round SIMON64. Additionally, we enhance the outcomes of the practical key recovery attack on SPECK large-state members, which amplifies the success rate of the key recovery attack in comparison to existing results.

改进的深度学习辅助密钥恢复框架：大状态分组密码的应用

[1]Baksi A, 2022. Machine learning-assisted differential distinguishers for lightweight ciphers. In: Baksi A (Ed.), Classical and Physical Security of Symmetric Key Cryptographic Algorithms. Springer, Singapore, p.141-162.

[2]Bao ZZ, Guo J, Liu MC, et al., 2022. Enhancing differential-neural cryptanalysis. 28^th Int Conf on the Theory and Application of Cryptology and Information Security, p.318-347.

[3]Beaulieu R, Shors D, Smith J, et al., 2015. The SIMON and SPECK lightweight block ciphers. Proc 52^nd Annual Design Automation Conf, Article 175.

[4]Bellini E, Rossi M, 2021. Performance comparison between deep learning-based and conventional cryptographic distinguishers. Proc Computing Conf on Intelligent Computing, p.681-701.

[5]Biham E, 1994. New types of cryptanalytic attacks using related keys. J Cryptol, 7(4):229-246.

[6]Biham E, Shamir A, 1993. Differential cryptanalysis of the full 16-round DES. 12^th Annual Int Cryptology Conf on Advances in Cryptology, p.487-496.

[7]Chen Y, Yu HB, 2021. A new neural distinguisher model considering derived features from multiple ciphertext pairs. Comput J, Article 310.

[8]Chen Y, Bao ZZ, Shen YT, et al., 2022. A deep learning aided key recovery framework for large-state block ciphers. Sci China Inform, 53(7):1348-1367 (in Chinese).

[9]Chen Y, Shen YT, Yu HB, 2023. Neural-aided statistical attack for cryptanalysis. Comput J, 66(10):2480-2498.

[10]Gohr A, 2019. Improving attacks on round-reduced Speck32/64 using deep learning. 39^th Annual Int Cryptology Conf on Advances in Cryptology, p.150-179.

[11]Hou ZZ, Ren JJ, Chen SZ, 2023. Practical attacks of round-reduced SIMON based on deep learning. Comput J, 66(10):2517-2534.

[12]Jain A, Kohli V, Mishra G, 2020. Deep learning based differential distinguisher for lightweight cipher PRESENT. https://eprint.iacr.org/2020/846

[13]Kingma DP, Ba J, 2017. Adam: a method for stochastic optimization.

[14]Knudsen LR, 1991. Cryptanalysis of LOKI. Int Conf on the Theory and Application of Cryptology, p.22-35.

[15]Zhang L, Wang ZL, Wang BY, 2022. Improving differential-neural cryptanalysis with inception blocks. https://dblp.org/rec/journals/iacr/zhangWW22.html