Improving Differential-Neural Cryptanalysis

Liu Zhang; Zilong Wang; Baocang Wang

doi:10.62056/ay11wa3y6

Improving Differential-Neural Cryptanalysis

Authors

Liu Zhang, Zilong Wang, Baocang Wang

Liu Zhang

School of Cyber Engineering, Xidian University, Xi'an, China
State Key Laboratory of Cryptology, P.O.Box 5159, Beijing, China
liuzhang at stu dot xidian dot edu dot cn

Zilong Wang

School of Cyber Engineering, Xidian University, Xi'an, China
State Key Laboratory of Cryptology, P.O.Box 5159, Beijing, China
zlwang at xidian dot edu dot cn

Baocang Wang

State Key Laboratory of Integrated Service Networks, Xidian University, Xi'an, China
bcwang79 at aliyun dot com

Keywords: Differential-Neural Distinguisher Inception Module Speck Simon Knowledge Distillation Key Recovery Attack

Abstract

Our first objective is to enhance the capabilities of differential-neural distinguishers by applying more deep-learning techniques, focusing on handling more rounds and improving accuracy. Inspired by the Inception Block in GoogLeNet, we adopted a design that uses multiple parallel convolutional layers with varying kernel sizes before the residual block to capture multi-dimensional information. Additionally, we expanded the convolutional kernels in the residual blocks, enlarging the network's receptive field. In the case of Speck32/64, our efforts yield accuracy improvements in rounds 6, 7, and 8, enabling the successful training of a 9-round differential-neural distinguisher. As for Simon32/64, we developed a differential-neural distinguisher capable of effectively handling 12 rounds while achieving noteworthy accuracy enhancements in rounds 9, 10, and 11.

Additionally, we utilized neutral bits to ensure the required data distribution for launching a successful key recovery attack when using multiple-ciphertext pairs as input for the neural network. Meanwhile, we redefined the formula for time complexity based on the differences in prediction speeds of the distinguisher between a single-core CPU and a GPU. Combining these various advancements allows us to considerably reduce the time and data complexity of key recovery attacks on 13-round Speck32/64. Furthermore, we used knowledge distillation techniques to reduce the model size, accelerating the distinguisher's prediction speed and reducing the time complexity. In particular, we achieved a successful 14-round key recovery attack by exhaustively guessing a 1-round subkey. For Simon32/64, we accomplished a 17-round key recovery attack for the first time and reduced the time complexity of the 16-round key recovery attack.

References

[AL13]

Hoda A. Alkhzaimi and Martin M. Lauridsen. Cryptanalysis of the SIMON Family of Block Ciphers. Cryptology ePrint Archive, Paper 2013/543. 2013.

Google Scholar ePrint

[ALLW14]

Farzaneh Abed, Eik List, Stefan Lucks, and Jakob Wenzel. Differential Cryptanalysis of Round-Reduced Simon and Speck. In Carlos Cid and Christian Rechberger, editors, Fast Software Encryption - 21st International Workshop, FSE 2014, London, UK, March 3-5, 2014. Revised Selected Papers, volume 8540 of Lecture Notes in Computer Science, pages 525–545. 2014. Springer. DOI: 10.1007/978-3-662-46706-0_27

Google Scholar ePrint

[BC04]

Eli Biham and Rafi Chen. Near-Collisions of SHA-0. In Matthew K. Franklin, editor, Advances in Cryptology - CRYPTO 2004, 24th Annual International CryptologyConference, Santa Barbara, California, USA, August 15-19, 2004, Proceedings, volume 3152 of Lecture Notes in Computer Science, pages 290–305. 2004. Springer. DOI: 10.1007/978-3-540-28628-8_18

Google Scholar ePrint

[BdST⁺23]

Alex Biryukov, Luan Cardoso dos Santos, Je Sen Teh, Aleksei Udovenko, and Vesselin Velichkov. Meet-in-the-Filter and Dynamic Counting with Applications to Speck. In Mehdi Tibouchi and Xiaofeng Wang, editors, Applied Cryptography and Network Security - 21st International Conference, ACNS 2023, Kyoto, Japan, June 19-22, 2023, Proceedings, Part I, volume 13905 of Lecture Notes in Computer Science, pages 149–177. 2023. Springer. DOI: 10.1007/978-3-031-33488-7_6

Google Scholar ePrint

[BGL⁺22]

Zhenzhen Bao, Jian Guo, Meicheng Liu, Li Ma, and Yi Tu. Enhancing Differential-Neural Cryptanalysis. In Shweta Agrawal and Dongdai Lin, editors, Advances in Cryptology - ASIACRYPT 2022 - 28th International Conference on the Theory and Application of Cryptology and Information Security, Taipei, Taiwan, December 5-9, 2022, Proceedings, Part I, volume 13791 of Lecture Notes in Computer Science, pages 318–347. 2022. Springer. DOI: 10.1007/978-3-031-22963-3_11

Google Scholar ePrint

[BGPT21]

Adrien Benamira, David Gérault, Thomas Peyrin, and Quan Quan Tan. A Deeper Look at Machine Learning-Based Cryptanalysis. In Anne Canteaut and François-Xavier Standaert, editors, Advances in Cryptology - EUROCRYPT 2021 - 40th Annual International Conference on the Theory and Applications of Cryptographic Techniques, Zagreb, Croatia, October 17-21, 2021, Proceedings, Part I, volume 12696 of Lecture Notes in Computer Science, pages 805–835. 2021. Springer. DOI: 10.1007/978-3-030-77870-5_28

Google Scholar ePrint

[BRV14]

Alex Biryukov, Arnab Roy, and Vesselin Velichkov. Differential Analysis of Block Ciphers SIMON and SPECK. In Carlos Cid and Christian Rechberger, editors, Fast Software Encryption - 21st International Workshop, FSE 2014, London, UK, March 3-5, 2014. Revised Selected Papers, volume 8540 of Lecture Notes in Computer Science, pages 546–570. 2014. Springer. DOI: 10.1007/978-3-662-46706-0_28

Google Scholar ePrint

[BSS⁺15]

Ray Beaulieu, Douglas Shors, Jason Smith, Stefan Treatman-Clark, Bryan Weeks, and Louis Wingers. The SIMON and SPECK lightweight block ciphers. In Proceedings of the 52nd Annual Design Automation Conference, San Francisco, CA, USA, June 7-11, 2015, pages 175:1–175:6. 2015. ACM. DOI: 10.1145/2744769.2747946

Google Scholar ePrint

[CSYY23]

Yi Chen, Yantian Shen, Hongbo Yu, and Sitong Yuan. A New Neural Distinguisher Considering Features Derived From Multiple Ciphertext Pairs. Comput. J., 66(6):1419–1433, 2023. DOI: 10.1093/COMJNL/BXAC019

Google Scholar ePrint

[Din14]

Itai Dinur. Improved Differential Cryptanalysis of Round-Reduced Speck. In Antoine Joux and Amr M. Youssef, editors, Selected Areas in Cryptography - SAC 2014 - 21st International Conference, Montreal, QC, Canada, August 14-15, 2014, Revised Selected Papers, volume 8781 of Lecture Notes in Computer Science, pages 147–164. 2014. Springer. DOI: 10.1007/978-3-319-13051-4_9

Google Scholar ePrint

[FLW⁺23]

Zhuohui Feng, Ye Luo, Chao Wang, Qianqian Yang, Zhiquan Liu, and Ling Song. Improved Differential Cryptanalysis on SPECK Using Plaintext Structures. In Leonie Simpson and Mir Ali Rezazadeh Baee, editors, Information Security and Privacy - 28th Australasian Conference, ACISP 2023, Brisbane, QLD, Australia, July 5-7, 2023, Proceedings, volume 13915 of Lecture Notes in Computer Science, pages 3–24. 2023. Springer. DOI: 10.1007/978-3-031-35486-1_1

Google Scholar ePrint

[GLN22]

Aron Gohr, Gregor Leander, and Patrick Neumann. An Assessment of Differential-Neural Distinguishers. Cryptology ePrint Archive, Paper 2022/1521. 2022.

Google Scholar ePrint

[Goh19]

Aron Gohr. Improving Attacks on Round-Reduced Speck32/64 Using Deep Learning. In Alexandra Boldyreva and Daniele Micciancio, editors, Advances in Cryptology - CRYPTO 2019 - 39th Annual International Cryptology Conference, Santa Barbara, CA, USA, August 18-22, 2019, Proceedings, Part II, volume 11693 of Lecture Notes in Computer Science, pages 150–179. 2019. Springer. DOI: 10.1007/978-3-030-26951-7_6

Google Scholar ePrint

[HLvdMW17]

Gao Huang, Zhuang Liu, Laurens van der Maaten, and Kilian Q. Weinberger. Densely Connected Convolutional Networks. In 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, July 21-26, 2017, pages 2261–2269. 2017. IEEE Computer Society. DOI: 10.1109/CVPR.2017.243

Google Scholar ePrint

[HSS18]

Jie Hu, Li Shen, and Gang Sun. Squeeze-and-Excitation Networks. In 2018 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, UT, USA, June 18-22, 2018, pages 7132–7141. 2018. Computer Vision Foundation / IEEE Computer Society. DOI: 10.1109/CVPR.2018.00745

Google Scholar ePrint

[HZRS16]

Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep Residual Learning for Image Recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV, USA, June 27-30, 2016, pages 770–778. 2016. IEEE Computer Society. DOI: 10.1109/CVPR.2016.90

Google Scholar ePrint

[KB15]

Diederik P. Kingma and Jimmy Ba. Adam: A Method for Stochastic Optimization. In Yoshua Bengio and Yann LeCun, editors, 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. 2015.

Google Scholar ePrint

[SHY16]

Ling Song, Zhangjie Huang, and Qianqian Yang. Automatic Differential Analysis of ARX Block Ciphers with Application to SPECK and LEA. In Joseph K. Liu and Ron Steinfeld, editors, Information Security and Privacy - 21st Australasian Conference, ACISP 2016, Melbourne, VIC, Australia, July 4-6, 2016, Proceedings, Part II, volume 9723 of Lecture Notes in Computer Science, pages 379–394. 2016. Springer. DOI: 10.1007/978-3-319-40367-0_24

Google Scholar ePrint

[SLJ⁺15]

Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott E. Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. Going deeper with convolutions. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, MA, USA, June 7-12, 2015, pages 1–9. 2015. IEEE Computer Society. DOI: 10.1109/CVPR.2015.7298594

Google Scholar ePrint

[WWJZ18]

Ning Wang, Xiaoyun Wang, Keting Jia, and Jingyuan Zhao. Differential attacks on reduced SIMON versions with dynamic key-guessing techniques. Sci. China Inf. Sci., 61(9):098103:1–098103:3, 2018. DOI: 10.1007/S11432-017-9231-5

Google Scholar ePrint

PDF Open access

DOI: https://doi.org/10.62056/ay11wa3y6

History

Submitted: 2024-07-04
Accepted: 2024-09-02
Published: 2024-10-07

How to cite

Liu Zhang, Zilong Wang, and Baocang Wang, Improving Differential-Neural Cryptanalysis. IACR Communications in Cryptology, vol. 1, no. 3, Oct 07, 2024, doi: 10.62056/ay11wa3y6.

License

Copyright is held by the author(s)

This work is licensed under a Creative Commons Attribution (CC BY) license.

@article{CiC-1-3-13,
    author = "Zhang, Liu and Wang, Zilong and Wang, Baocang",
    journal = "{IACR} {C}ommunications in {C}ryptology",
    publisher = "{I}nternational {A}ssociation for {C}ryptologic {R}esearch",
    title = "Improving Differential-Neural Cryptanalysis",
    volume = "1",
    number = "3",
    date = "2024-10-07",
    year = "2024",
    issn = "3006-5496",
    doi = "10.62056/ay11wa3y6"
}

TY - JOUR
AU - Zhang, Liu
AU - Wang, Zilong
AU - Wang, Baocang
PY  - 2024
TI  - Improving Differential-Neural Cryptanalysis
JF  - IACR Communications in Cryptology
JA  - CIC
VL  - 1
IS  - 3
DO  - 10.62056/ay11wa3y6
UR  - https://doi.org/10.62056/ay11wa3y6
AB  - <p>     Our first objective is to enhance the capabilities of differential-neural distinguishers by applying more deep-learning techniques, focusing on handling more rounds and improving accuracy. Inspired by the Inception Block in GoogLeNet, we adopted a design that uses multiple parallel convolutional layers with varying kernel sizes before the residual block to capture multi-dimensional information. Additionally, we expanded the convolutional kernels in the residual blocks, enlarging the network's receptive field.     In the case of Speck32/64, our efforts yield accuracy improvements in rounds 6, 7, and 8, enabling the successful training of a 9-round differential-neural distinguisher. As for Simon32/64, we developed a differential-neural distinguisher capable of effectively handling 12 rounds while achieving noteworthy accuracy enhancements in rounds 9, 10, and 11.</p><p>    Additionally, we utilized neutral bits to ensure the required data distribution for launching a successful key recovery attack when using multiple-ciphertext pairs as input for the neural network.     Meanwhile, we redefined the formula for time complexity based on the differences in prediction speeds of the distinguisher between a single-core CPU and a GPU.     Combining these various advancements allows us to considerably reduce the time and data complexity of key recovery attacks on 13-round Speck32/64.     Furthermore, we used knowledge distillation techniques to reduce the model size, accelerating the distinguisher's prediction speed and reducing the time complexity.     In particular, we achieved a successful 14-round key recovery attack by exhaustively guessing a 1-round subkey. For Simon32/64, we accomplished a  17-round key recovery attack for the first time and reduced the time complexity of the 16-round key recovery attack. </p>
ER  -

Liu Zhang, Zilong Wang, and Baocang Wang, Improving Differential-Neural Cryptanalysis. IACR Communications in Cryptology, vol. 1, no. 3, Oct 07, 2024, doi: 10.62056/ay11wa3y6.