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

• 学术论文 • 上一篇    下一篇

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. 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. 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:

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 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 1 10 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: 

No Suggested Reading articles found!