基于深度学习的CRAFT算法差分分析

沈璇; 沈霞民; 卜予彤; 代威; 黄金杰; 郭连城

doi:10.11959/j.issn.2096-109x.AQ25290

您当前的位置：

首页 >

文章列表页 >

基于深度学习的CRAFT算法差分分析

学术论文 | 更新时间：2026-07-15

- 基于深度学习的CRAFT算法差分分析
- Differential analysis of the CRAFT algorithm based on deep learning
- 网络与信息安全学报 2026年12卷第3期页码：122-134
- 作者机构：
  
  1.信息支援部队工程大学，湖北武汉 430010
  2.国防科技大学，湖南长沙 410073
- 作者简介：
  
  沈璇（1990- ），男，博士，信息支援部队工程大学副教授，主要研究方向为对称密码的设计与分析。
  沈霞民（1993− ），男，国防科技大学硕士生，主要研究方向为对称密码的设计与分析。
  卜予彤（2001- ），女，国防科技大学硕士生，主要研究方向为对称密码的设计与分析。
  代威（1991- ），男，硕士，信息支援部队工程大学讲师，主要研究方向为网络安全防护。
  黄金杰（1990- ），男，信息支援部队工程大学讲师，主要研究方向为信息安全。
  郭连城（1980- ），男，信息支援部队工程大学副教授，主要研究方向为软件漏洞分析、软件加解密。
- 基金信息：
  
  国家自然科学基金资助项目(62272470)
- DOI：10.11959/j.issn.2096-109x.AQ25290
  中图分类号： TN918
- 收稿：2025-12-07，
  
  修回：2026-05-18，
  
  纸质出版：2026-06-25
- 稿件说明：
移动端阅览
沈璇, 沈霞民, 卜予彤, 等. 基于深度学习的CRAFT算法差分分析[J]. 网络与信息安全学报, 2026, 12(3): 122-134.

Shen Xuan, Shen Xiamin, Bu Yutong, et al. Differential analysis of the CRAFT algorithm based on deep learning[J]. Chinese Journal of Network and Information Security, 2026, 12(3): 122-134.
沈璇, 沈霞民, 卜予彤, 等. 基于深度学习的CRAFT算法差分分析[J]. 网络与信息安全学报, 2026, 12(3): 122-134. DOI： 10.11959/j.issn.2096-109x.AQ25290.

Shen Xuan, Shen Xiamin, Bu Yutong, et al. Differential analysis of the CRAFT algorithm based on deep learning[J]. Chinese Journal of Network and Information Security, 2026, 12(3): 122-134. DOI： 10.11959/j.issn.2096-109x.AQ25290.

摘要

基于深度学习的分组密码差分分析已成为当前密码学领域研究热点，为研究不同数据格式对神经网络差分区分器性能的影响，结合单差分、多差分区分器数据格式特点，利用CRAFT算法S盒组

件具有差分传播不确定性的特性，设计提出“单差分+S盒中间状态”区分器数据格式，通过残差神经网络模型分别训练获得3类数据格式对应的CRAFT算法5~7轮神经网络差分区分器。对于5轮神经网络差分区分器，3类区分器准确率均在97%以上；对于6轮神经网络差分区分器，“单差分+S盒中间状态”区分器准确率为96.14%，明显优于单差分区分器的61.39%和多差分区分器的72.78%；对于7轮神经网络差分区分器，3类区分器准确率略高于50%。其中，单差分和“单差分+S盒中间状态”区分器对应模型训练数据复杂度为

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007348&type=

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007359&type=

6.26533365

2.45533323

，多差分区分器对应模型训练数据复杂度为

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007337&type=

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007349&type=

6.26533365

2.45533323

。进一步利用6轮“单差分+S盒中间状态”区分器对7轮CRAFT算法进行部分密钥恢复攻击，当选择明密文对数量为10 000个时，在100次攻击实验中成功率可达93%，一次完整的密码分析攻击时间复杂度为

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007353&type=

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007364&type=

6.26533365

2.45533323

。

Abstract

The differential analysis of block ciphers based on deep learning has become a hot research topic in the field of cryptography. To investigate the impact of different data formats on the performance of neural network differential distinguisher

by combining the characteristics of single-differential and multi-differential distinguisher data formats

and taking advantage of the differential propagation uncertainty of the S-box component in the CRAFT

a “single-differential + S-box intermediate state” distinguisher data format was designed and proposed. Through residual neural network (ResNet) models

three types of data formats corresponding to the 5-7 rounds of neural network differential distinguisher of the CRAFT were obtained. For the 5-round neural network differential distinguisher

the accuracy of all three discriminators was above 97%. For the 6-round neural network differential distinguisher

the accuracy of the “single-differential + S-box intermediate state” distinguisher was 96.14%

significantly better than the 61.39% of the single-differential distinguisher and the 72.78% of the multi-differential distinguisher. For the 7-round neural network differential distinguisher

the accuracy of the three distinguishers was slightly above 50%. Among them

the training data complexity of the model corresponding to the single-differential distinguisher was

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007368&type=

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007354&type=

6.26533365

2.45533323

while the training data complexity of the multi-differential and “single-differential + S-box intermediate state” distinguisher was

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007379&type=

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007357&type=

6.26533365

2.45533323

. Further

by using a 6-round “single difference + S-box intermediate state” distinguisher to conduct partial key recovery attacks on the 7-round CRAFT

when the number of plaintext-ciphertext pairs selected was 10 000

the success rate can reach 93% in 100 attack experiments

the time complexity of a complete cryptanalysis attack was

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007381&type=

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=113007380&type=

6.26533365

2.45533323

关键词

Keywords

references

钟悦 , 谷杰铭 , 曹洪林 . 轻量级分组密码算法综述 [J ] . 计算机科学 , 2023 , 50 ( 9 ): 3 - 15 .

Zhong Y , Gu J M , Cao H L . Survey of lightweight block cipher [J ] . Computer Science , 2023 , 50 ( 9 ): 3 - 15 .

Beierle C , Jean J , Kölbl S , et al . The SKINNY family of block ciphers and its low-latency variant MANTIS [C ] // Advances in Cryptology-CRYPTO 2016 . Berlin : Springer , 2016 : 123 - 153 .

Beierle C , Leander G , Moradi A , et al . CRAFT: lightweight tweakable block cipher with efficient protection against DFA attacks [J ] . IACR Transactions on Symmetric Cryptology , 2019 : 5 - 45 .

Beaulieu R , Shors D , Smith J , et al . The SIMON and SPECK families of lightweight block ciphers [J ] . IACR Cryptology EPrint Archive , 2013 , 2013 : 404 .

沈璇 , 刘国强 , 孙兵 , 等 . 两类动态密码结构抵抗不可能差分和零相关线性能力评估 [J ] . 电子学报 , 2024 , 52 ( 3 ): 709 - 718 .

Shen X , Liu G Q , Sun B , et al . Security evaluation against impossible differential cryptanalysis and zero correlation linear cryptanalysis for two dynamic cryptographic structures [J ] . Acta Electronica Sinica , 2024 , 52 ( 3 ): 709 - 718 .

成磊 , 沈璇 , 任传伦 . 广义类CLEFIA动态密码结构抵抗差分和线性密码分析的安全性评估 [J ] . 电子学报 , 2024 , 52 ( 8 ): 2571 - 2580 .

Cheng L , Shen X , Ren C L . Security evaluation of generalized CLEFIA-like dynamic cipher structures against differential and linear cryptanalysis [J ] . Acta Electronica Sinica , 2024 , 52 ( 8 ): 2571 - 2580 .

Gohr A . Improving attacks on Speck32/64 using deep learning [J ] . IACR Cryptology EPrint Archive , 2019 , 2019 : 37 .

Chen Y , Shen Y T , Yu H B , et al . A new neural distinguisher considering features derived from multiple ciphertext pairs [J ] . The Computer Journal , 2023 , 66 ( 6 ): 1419 - 1433 .

杨小雪 , 陈杰 . 几类密码算法的神经网络差分区分器的改进 [J ] . 西安电子科技大学学报(自然科学版) , 2024 , 51 ( 1 ): 210 - 222 .

Yang X X , Chen J . Improvement of the neural distinguishers of several ciphers [J ] . Journal of Xidian University (Natural Science) , 2024 , 51 ( 1 ): 210 - 222 .

段明 , 付超辉 , 魏强 , 等 . 基于深度学习的多差分神经网络区分器及其应用 [J ] . 信息工程大学学报 , 2021 , 22 ( 3 ): 347 - 350 .

Duan M , Fu C H , Wei Q , et al . Multi differential attack based on deep learning and its application [J ] . Journal of Information Engineering University , 2021 , 22 ( 3 ): 347 - 350 .

蒋若怡 , 韦永壮 , 王慧娇 . 基于深度学习的差分神经区分器求解方法 [J ] . 计算机工程与设计 , 2023 , 44 ( 6 ): 1629 - 1634 .

Jiang R Y , Wei Y Z , Wang H J . Method for solving differential neural distinguishers based on deep learning [J ] . Computer Engineering and Design , 2023 , 44 ( 6 ): 1629 - 1634 .

王慧娇 , 张鑫 , 韦永壮 , 等 . 基于深度学习的SM4密码算法新型区分器 [J ] . 通信学报 , 2023 , 44 ( 7 ): 171 - 184 .

Wang H J , Zhang X , Wei Y Z , et al . Novel distinguisher for SM4 cipher algorithm based on deep learning [J ] . Journal on Communications , 2023 , 44 ( 7 ): 171 - 184 .

侯泽洲 , 陈少真 , 任炯炯 . 深度学习在分组密码差分区分器上的研究应用 [J ] . 软件学报 , 2022 , 33 ( 5 ): 1893 - 1906 .

Hou Z Z , Chen S Z , Ren J J . Research and application of deep learning on differential distinguisher of block cipher [J ] . Journal of Software , 2022 , 33 ( 5 ): 1893 - 1906 .

栗琳轲 , 陈杰 , 刘君 . 神经网络差分区分器的改进方案与应用 [J ] . 西安电子科技大学学报(自然科学版) , 2025 , 52 ( 1 ): 196 - 214 .

Li L K , Chen J , Liu J . Improved schemes and applications of the neural network differential distinguisher [J ] . Journal of Xidian University (Natural Science) , 2025 , 52 ( 1 ): 196 - 214 .

Hadipour H , Sadeghi S , Niknam M M , et al . Comprehensive security analysis of CRAFT [J ] . IACR Transactions on Symmetric Cryptology , 2020 : 290 - 317 .

Liu H M , Zhang W Y , Zhang J J , et al . Clustering of differentials in CRAFT with correlation matrices [J ] . International Journal of Intelligent Systems , 2022 , 37 ( 12 ): 12113 - 12134 .

蒋若怡 . 分组密码的差分类自动化分析方法研究 [D ] . 桂林 : 桂林电子科技大学 , 2022 .

Jiang R Y . Research on automatic analysis methods of differential classification in block ciphers [D ] . Gulin : Guilin University of Elect-Ronic Technology , 2022 .

Ebrahimi M A , Ahmadian Z . New automatic search method for truncated-differential characteristics application to midori, SKINNY and CRAFT [J ] . The Computer Journal , 2020 , 63 ( 12 ): 1813 - 1825 .

Hadipour H , Bagheri N , Song L . Improved rectangle attacks on SKINNY and CRAFT [J ] . IACR Transactions on Symmetric Cryptology , 2021 : 140 - 198 .

Hadipour H , Sadeghi S , Eichlseder M . Finding the impossible: automated search for full impossible-differential, zero-correlation, and integral attacks [C ] // Advances in Cryptology-EUROCRYPT 2023 . Berlin : Springer , 2023 : 128 - 157 .

李超 , 孙兵 , 李瑞林 . 分组密码的攻击方法与实例分析 [M ] . 北京 : 科学出版社 , 2010 .

LI C , SUN B , LI R L . Attack methods and case analysis of block ciphers [M ] . Beijing : Science Press , 2010 .

黄锦坤 , 黄远航 , 黄文敏 , 等 . 基于跨流注意力增强中心差分卷积网络的CG图像检测 [J ] . 网络与信息安全学报 , 2024 , 10 ( 6 ): 96 - 108 .

Huang J K , Huang Y H , Huang W M , et al . Cross-stream attention enhanced central difference convolutional network for CG image detection [J ] . Chinese Journal of Network and Information Security , 2024 , 10 ( 6 ): 96 - 108 .

刘宇翔 , 莫秀良 , 宫良一 , 等 . 基于时空特征融合深度学习的NXNSAttack流量识别方法 [J ] . 网络与信息安全学报 , 2025 , 11 ( 4 ): 173 - 188 .

Liu Y X , Mo X L , Gong L Y , et al . NXNSAttack traffic detection method based on spatio-temporal feature fusion deep learning [J ] . Chinese Journal of Network and Information Security , 2025 , 11 ( 4 ): 173 - 188 .

夏锐琪 , 李曼曼 , 陈少真 . 基于机器学习的分组密码结构识别 [J ] . 网络与信息安全学报 , 2023 , 9 ( 3 ): 79 - 89 .

Xia R Q , Li M M , Chen S Z . Identification on the structures of block ciphers using machine learning [J ] . Chinese Journal of Network and Information Security , 2023 , 9 ( 3 ): 79 - 89 .

He K M , Zhang X Y , Ren S Q , et al . Deep residual learning for image recognition [C ] // Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE Press , 2016 : 770 - 778 .

Benamira A , Gerault D , Peyrin T , et al . A deeper look at machine learning-based cryptanalysis [C ] // Advances in Cryptology-EUROCRYPT 2021 . Berlin : Springer , 2021 : 805 - 835 .

Bao Z Z , Lu J Y , Yao Y R , et al . More insight on deep learning-aided cryptanalysis [C ] // Advances in Cryptology-ASIACRYPT 2023 . Berlin : Springer , 2023 : 436 - 467 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于双路特征提取和自适应融合的钓鱼网站检测

DNS异常检测：大规模实验分析

FL-IoV：基于联邦学习的综合车联网入侵检测方法

基于分层结构的多任务对抗样本归因方法

彩色图像隐写分析研究进展