基于深度学习的随机性检验策略研究

doi:10.11959/j.issn.1000-436x.2023111

通信学报 ›› 2023, Vol. 44 ›› Issue (6): 23-33.doi: 10.11959/j.issn.1000-436x.2023111

基于深度学习的随机性检验策略研究

陈东昱¹^,², 陈华¹, 范丽敏¹, 付一方¹^,², 王舰¹^,²

¹ 中国科学院软件研究所可信计算与信息保障实验室，北京 100190
² 中国科学院大学，北京 100049

修回日期:2023-06-12 出版日期:2023-06-25 发布日期:2023-06-01
作者简介:陈东昱（1989- ），男，山东淄博人，中国科学院软件研究所博士生，主要研究方向为随机数发生器设计与随机性统计检验
陈华（1976- ），女，山东日照人，博士，中国科学院软件研究所正高级工程师、博士生导师，主要研究方向为侧信道分析与防护、密码检测
范丽敏（1978- ），女，内蒙古赤峰人，博士，中国科学院软件研究所高级工程师，主要研究方向为随机性检验、密码检测及侧信道分析与防护
付一方（1997- ），男，辽宁凌源人，中国科学院软件研究所硕士生，主要研究方向为随机数发生器设计与随机性统计检验
王舰（1998- ），男，山东临沂人，中国科学院软件研究所博士生，主要研究方向为侧信道分析与防护
基金资助:
国家重点研发计划基金资助项目(2020YFA0309704)

Research on test strategy for randomness based on deep learning

Dongyu CHEN¹^,², Hua CHEN¹, Limin FAN¹, Yifang FU¹^,², Jian WANG¹^,²

¹ Trusted Computing and Information Assurance Laboratory, Institute of Software, Chinese Academy of Sciences, Beijing 100190, China
² University of Chinese Academy of Sciences, Beijing 100049, China

Revised:2023-06-12 Online:2023-06-25 Published:2023-06-01
Supported by:
The National Key Research and Development Program of China(2020YFA0309704)

摘要/Abstract

摘要：

为了获得更好的检验效果，对基于深度学习的随机性检验策略进行了研究，包括 2021 年欧密会提出的批均化策略和数据单元大小的选择策略。通过给出基于深度学习方法的随机性统计检验模型，理论推导得到2个检验策略的统计量分布和检验势表达，并指出：1) 批均化策略虽然能够提升模型预测准确率，但在统计上容易造成第二类错误概率的增大，反而降低了检验势；2) 一般情况下深度学习模型的数据单元越小，取得的检验势越高。基于以上认识，提出了一种新的比特级深度学习模型用于随机性统计检验。该模型应用于线性同余发生器（LCG）算法，相比之前工作，参数量减少至 $\frac{1}{80}$ ，取得预测优势所需数据减少了50%以上；拓展应用于5～7轮Speck算法获得了明显的预测优势，与Gohr模型相比，参数量减少至 $\frac{1}{10}$ ～ $\frac{1}{20}$ 。

关键词: 深度学习, 随机性, 统计检验, 随机数发生器, Speck, 线性同余发生器, 批均化策略

Abstract:

In order to achieve better test performance, researches on the randomness test strategies based on deep learning were conducted, including the batch average strategy proposed by EUROCRYPT 2021 and the selection strategy for data unit size.By introducing the randomness statistical test model based on deep learning methods, the statistical distribution and test power expression of two test strategies were theoretically derived, and it was pointed out that: (i) the batch average strategy could amplify the prediction accuracy of the model, but it was prone to an increase in the probability of the second type of error in statistics, instead reducing the statistical test power; (ii) the smaller data units of the deep model generally obtained the more powerful statistical tests.Based on the above understanding, a new bit-level deep learning model was proposed for randomness statistical tests, which gained the advantage of prediction with 80 times fewer parameters and 50% samples, compared with the previous work on linear congruent generator (LCG) algorithm, and achieved significant prediction advantages with 10～20 times fewer parameters by extending the model to apply to 5～7 rounds of Speck, compared with the model proposed by Gohr.

Key words: deep learning, randomness, statistical test, random number generator, Speck, LCG, batch average strategy

中图分类号:

TN918.1

陈东昱, 陈华, 范丽敏, 付一方, 王舰. 基于深度学习的随机性检验策略研究[J]. 通信学报, 2023, 44(6): 23-33.

Dongyu CHEN, Hua CHEN, Limin FAN, Yifang FU, Jian WANG. Research on test strategy for randomness based on deep learning[J]. Journal on Communications, 2023, 44(6): 23-33.

图/表 9

图1

图2

表1

图3

图4

图5

图6

图7

图8

参考文献 20

[1]	BOYAR J . Inferring sequences produced by a linear congruential generator missing low-order bits[J]. Journal of Cryptology, 1989,1(3): 177-184.
[2]	MATSUMOTO M , NISHIMURA T . Mersenne twister:a 623- dimensionally equidistributed uniform pseudo-random number generator[J]. ACM Transactions on Modeling and Computer Simulation, 1998,8(1): 3-30.
[3]	RUKHIN A , SOTO J , NECHVATAL J . A statistical test suite for random and pseudorandom number generators for cryptographic applications[R]. 2001.
[4]	KILLMANN W , SCHINDLER W . Functionality classes and evaluation methodology for true (physical) random number generators:AIS 31[S].(2001-09-25).
[5]	国家密码管理局. 随机性检测规范:GM/T 0005—2021[S]. 北京:中国标准出版社, 2021.
	State Cryptography Administration. Randomness testing specifications:GM/T 0005—2021[S. Beijing:Standards Press of China, 2021.
[6]	SULAK F . Statistical analysis of block ciphers and hash functions[D]. Ankara:Middle East Technical University, 2011.
[7]	ALANI M M . Applications of machine learning in cryptography:a survey[C]// Proceedings of the 3rd International Conference on Cryptography,Security and Privacy. New York:ACM Press, 2019: 23-27.
[8]	GOHR A . Improving attacks on round-reduced speck32/64 using deep learning[C]// Proceedings of Annual International Cryptology Conference. Berlin:Springer, 2019: 150-179.
[9]	BENAMIRA A , GERAULT D , PEYRIN T ,et al. A deeper look at machine learning-based cryptanalysis[C]// Proceedings of Annual International Conference on the Theory and Applications of Cryptographic Techniques. Berlin:Springer, 2021: 805-835.
[10]	BAO Z Z , GUO J , LIU M C ,et al. Enhancing differential-neural cryptanalysis[C]// Proceedings of International Conference on the Theory and Application of Cryptology and Information Security. Berlin:Springer, 2022: 318-347.
[11]	PATERSON K , POETTERING B , SCHULDT J C N . Big bias hunting in Amazonia:large-scale computation and exploitation of RC4 biases[C]// Proceedings of International Conference on the Theory and Application of Cryptology and Information Security. Berlin:Springer, 2014: 398-419.
[12]	MISHRA G , GUPTA I , MURTHY S V S S N V G K ,et al. Deep learning based cryptanalysis of stream ciphers[J]. Defence Science Journal, 2021,71(4): 499-506.
[13]	SAVICKY P , ROBNIK-?IKONJA M . Learning random numbers:a MATLAB anomaly[J]. Applied Artificial Intelligence, 2008,22(3): 254-265.
[14]	FAN F L , WANG G . Learning from pseudo-randomness with an artificial neural network-does God play pseudo-dice?[J]. IEEE Access, 2018,6: 22987-22992.
[15]	FISCHER T . Testing cryptographically secure pseudo random number generators with artificial neural networks[C]// Proceedings of 2018 17th IEEE International Conference on Trust,Security and Privacy In Computing and Communications/ 12th IEEE International Conference on Big Data Science and Engineering (TrustCom/BigDataSE). Piscataway:IEEE Press, 2018: 1214-1223.
[16]	TRUONG N D , HAW J Y , ASSAD S M ,et al. Machine learning cryptanalysis of a quantum random number generator[J]. IEEE Transactions on Information Forensics and Security, 2019,14(2): 403-414.
[17]	LI C , ZHANG J G , SANG L X ,et al. Deep learning-based security verification for a random number generator using white chaos[J]. Entropy, 2020,22(10): 1134.
[18]	YANG J , ZHU S Y , CHEN T Y ,et al. Neural network based min-entropy estimation for random number generators[C]// Proceedings of International Conference on Security and Privacy in Communication Systems. Berlin:Springer, 2018: 231-250.
[19]	ZHU S Y , MA Y , LI X S ,et al. On the analysis and improvement of min-entropy estimation on time-varying data[J]. IEEE Transactions on Information Forensics and Security, 2020,15: 1696-1708.
[20]	AHMADZADEH E , KIM H , JEONG O ,et al. A novel dynamic attack on classical ciphers using an attention-based LSTM encoder-decoder model[J]. IEEE Access, 2021,9: 60960-60970.

检验项目			p
检验项目		5轮	6轮	7轮
SP800-2	频数	0.000 000	0.000 000	0.237 403
	块内频数	0.000 000	0.000 000	0.679 483
	游程	0.000 000	0.000 000	0.856 241
	最长1游程	0.000 000	0.806 792	0.465 166
	矩阵秩	0.000 000	0.346 675	0.644 905
	离散傅里叶	0.000 000	0.000 000	0.183 176
	非重叠模板匹配	0.000 000	0.000 000	0.945 836
	重叠模板匹配	0.000 000	0.452 246	0.956 599
	通用统计	0.000 000	0.000 136	0.618 025
	线性复杂度	0.639 372	0.026 867	0.464 382
	重叠子序列	0.000 000	0.042 517	0.954 304
	近似熵	0.000 000	0.000 000	0.981 640
	累加和	0.000 000	0.000 000	0.416 601
	随机游动	0.000 000	0.000 000	0.255 535
	随机游动变量	0.000 000	0.000 000	0.564 618
Gohr分类准确率		0.929 513	0.788 817	0.612 240

基于深度学习的随机性检验策略研究

Research on test strategy for randomness based on deep learning

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 20

相关文章 15

Metrics

推荐阅读 0

[1]	李荣鹏, 汪丙炎, 张宏纲, 赵志峰. 知识增强的语义通信接收端设计[J]. 通信学报, 2023, 44(6): 70-76.
[2]	马帅, 裴科, 祁华艳, 李航, 曹雯, 王洪梅, 熊海良, 李世银. 基于生成模型的地磁室内高精度定位算法研究[J]. 通信学报, 2023, 44(6): 211-222.
[3]	杨洁, 董标, 付雪, 王禹, 桂冠. 基于轻量化分布式学习的自动调制分类方法[J]. 通信学报, 2022, 43(7): 134-142.
[4]	杨秀璋, 彭国军, 李子川, 吕杨琦, 刘思德, 李晨光. 基于Bert和BiLSTM-CRF的APT攻击实体识别及对齐研究[J]. 通信学报, 2022, 43(6): 58-70.
[5]	廖勇, 王世义. 高速移动环境下基于RM-Net的大规模MIMO CSI反馈算法[J]. 通信学报, 2022, 43(5): 166-176.
[6]	廖育荣, 王海宁, 林存宝, 李阳, 方宇强, 倪淑燕. 基于深度学习的光学遥感图像目标检测研究进展[J]. 通信学报, 2022, 43(5): 190-203.
[7]	赵增华, 童跃凡, 崔佳洋. 基于域自适应的Wi-Fi指纹设备无关室内定位模型[J]. 通信学报, 2022, 43(4): 143-153.
[8]	廖勇, 程港, 李玉杰. 基于深度展开的大规模MIMO系统CSI反馈算法[J]. 通信学报, 2022, 43(12): 77-88.
[9]	段雪源, 付钰, 王坤, 李彬. 基于简单统计特征的LDoS攻击检测方法[J]. 通信学报, 2022, 43(11): 53-64.
[10]	霍俊彦, 邱瑞鹏, 马彦卓, 杨付正. 基于最邻近帧质量增强的视频编码参考帧列表优化算法[J]. 通信学报, 2022, 43(11): 136-147.
[11]	康海燕, 冀源蕊. 基于本地化差分隐私的联邦学习方法研究[J]. 通信学报, 2022, 43(10): 94-105.
[12]	张红霞, 王琪, 王登岳, 王奔. 基于深度学习的区块链蜜罐陷阱合约检测[J]. 通信学报, 2022, 43(1): 194-202.
[13]	晏燕, 丛一鸣, Adnan Mahmood, 盛权政. 基于深度学习的位置大数据统计发布与隐私保护方法[J]. 通信学报, 2022, 43(1): 203-216.
[14]	朱叶, 余宜林, 郭迎春. HRDA-Net：面向真实场景的图像多篡改检测与定位算法[J]. 通信学报, 2022, 43(1): 217-226.
[15]	杨晓元, 毕新亮, 刘佳, 黄思远. 结合图像加密与深度学习的高容量图像隐写算法[J]. 通信学报, 2021, 42(9): 96-105.