基于显著图的电磁信号对抗样本生成方法

周侠; 张剑; 李宁安

doi:10.12263/DZXB.20230125

您当前的位置：

首页 >

文章列表页 >

基于显著图的电磁信号对抗样本生成方法

学术论文 | 更新时间：2025-12-08

- 基于显著图的电磁信号对抗样本生成方法
- An Electromagnetic Signal Adversarial Examples Generation Method Based on Saliency Map
- 电子学报 2023年51卷第7期页码：1917-1928
- 作者机构：
  
  武汉数字工程研究所，湖北武汉 430205
- 作者简介：
  
  [ "周侠男，1996年11月出生于贵州省安顺市.现为武汉数字工程研究所在读硕士.研究方向为人工智能对抗攻防.E-mail: zhou_xia1110@163.com" ]
  [ "张剑（通讯作者）男，1979年1月出生于湖北省宜昌市.博士毕业于华中科技大学电信系.现为武汉数字工程研究所博士生导师、研究员，主要研究方向为人工智能、电子战." ]
  [ "李宁安男，1992年2月出生于河南省开封市. 现为武汉数字工程研究所高级工程师.主要研究方向为人工智能安全测评. E-mail: 1912665736@qq.com" ]
- 基金信息：
  
  国防科技173计划技术领域基金项目（A类）(A24011)
- DOI：10.12263/DZXB.20230125
  中图分类号： TP183
- 收稿：2023-02-13，
  
  修回：2023-05-08，
  
  纸质出版：2023-07-25
- 稿件说明：
移动端阅览
周侠,张剑,李宁安.基于显著图的电磁信号对抗样本生成方法[J].电子学报,2023,51(07):1917-1928.

ZHOU Xia,ZHANG Jian,LI Ning-an.An Electromagnetic Signal Adversarial Examples Generation Method Based on Saliency Map[J].ACTA ELECTRONICA SINICA,2023,51(07):1917-1928.
周侠,张剑,李宁安.基于显著图的电磁信号对抗样本生成方法[J].电子学报,2023,51(07):1917-1928. DOI： 10.12263/DZXB.20230125.

ZHOU Xia,ZHANG Jian,LI Ning-an.An Electromagnetic Signal Adversarial Examples Generation Method Based on Saliency Map[J].ACTA ELECTRONICA SINICA,2023,51(07):1917-1928. DOI： 10.12263/DZXB.20230125.

摘要

基于深度学习的电磁信号识别模型具有高效、准确和人工干预少的优点，然而其与传统神经网络模型一样容易受到对抗样本的影响.研究对抗样本对测试和提升模型的安全性和鲁棒性有着重要意义.为生成高质量电磁信号对抗样本，本文提出了基于雅可比显著图批量特征点攻击算法（Batch Points Jacobian-based Saliency Map Attack， BP-JSMA）.与传统雅可比显著图的攻击方法相比，BP-JSMA通过批量选取关键特征点能够更快生成对抗样本.此外，针对电磁信号数据的特点，增加自适应扰动限制，使得生成的对抗样本更具隐蔽性.在公开数据集的实验结果表明，与雅可比显著图攻击方法相比，BP-JSMA在生成速度方面提升了11倍，隐蔽性提升了10%；而与传统快速梯度符号攻击算法相比，攻击成功率提升了24%，隐蔽性提升了20%.

Abstract

The electromagnetic signal recognition model based on deep learning has the advantages of high efficiency

high accuracy

and less manual intervention. However

it is as susceptible to adversarial examples as traditional neural network models. Studying adversarial examples is important for testing and improving the security and robustness of neural network models. In order to generate high-quality electromagnetic signal adversarial examples

this paper proposed a batch feature points Jacobian-based saliency map attack method (BP-JSMA). Compared with the traditional Jacobian saliency map attack method

BP-JSMA can generate adversarial examples faster by selecting key feature points in a batch. In addition

according to the characteristics of electromagnetic signal data

adaptive disturbance limitation is proposed to make the generated adversarial examples more covert. Experimental results on public datasets show that compared to the Jacobian saliency map attack method

BP-JSMA improves generation speed by 11 times and concealment by 10%. Compared with traditional fast gradient sign attack method

the attack success rate has been improved by 24%

and the concealment has been improved by 20%.

关键词

Keywords

references

KRIZHEVSKY A , SUTSKEVER I , HINTON G E . ImageNet classification with deep convolutional neural networks [J]. Communications of the ACM , 2017 , 60 ( 6 ): 84 - 90 .

O'SHEA T J , CORGAN J , CLANCY T C . Convolutional radio modulation recognition networks [M]// Engineering Applications of Neural Networks . Cham : Springer International Publishing , 2016 : 213 - 226 .

O'SHEA T J , ROY T , CLANCY T C . Over-the-air deep learning based radio signal classification [J]. IEEE Journal of Selected Topics in Signal Processing , 2018 , 12 ( 1 ): 168 - 179 .

李钦 , 刘伟 , 牛朝阳 , 等 . 低信噪比下基于分裂EfficientNet网络的雷达信号调制方式识别 [J]. 电子学报 , 2023 , 51 ( 3 ): 675 - 686 .

LI Q , LIU W , NIU C Y , et al . Radar signal modulation recognition based on split EfficientNet under low signal-to-noise ratio [J]. Acta Electronica Sinica , 2023 , 51 ( 3 ): 675 - 686 . (in Chinese)

周鑫 , 何晓新 , 郑昌文 . 基于图像深度学习的无线电信号识别 [J]. 通信学报 , 2019 , 40 ( 7 ): 114 - 125 .

ZHOU X , HE X X , ZHENG C W . Radio signal recognition based on image deep learning [J]. Journal on Communications , 2019 , 40 ( 7 ): 114 - 125 . (in Chinese)

SZEGEDY C , ZAREMBA W , SUTSKEVER I , et al . Intriguing properties of neural networks [C]// 2014 the 2nd International Conference on Learning Representations . Banff : Conference Track Proceedings , 2014 : 1 - 10 .

GOODFELLOW I J , SHLENS J , SZEGEDY C . Explaining and harnessing adversarial examples [C]// 2015 the 3rd International Conference on Learning Representations . San Diego : Conference Track Proceedings , 2015 : 1 - 11 .

DONG Y P , LIAO F Z , PANG T Y , et al . Boosting adversarial attacks with momentum [C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 9185 - 9193 .

MADRY A , MAKELOV A , SCHMIDT L , et al . Towards deep learning models resistant to adversarial attacks [C]// 2018 the 6th International Conference on Learning Representations . Vancouver : Conference Track Proceedings , 2018 : 1 - 28 .

邹军华 , 段晔鑫 , 任传伦 , 等 . 基于噪声初始化、Adam-Nesterov方法和准双曲动量方法的对抗样本生成方法 [J]. 电子学报 , 2022 , 50 ( 1 ): 207 - 216 .

ZOU J H , DUAN Y X , REN C L , et al . Perturbation initialization, Adam-Nesterov and quasi-hyperbolic momentum for adversarial examples [J]. Acta Electronica Sinica , 2022 , 50 ( 1 ): 207 - 216 . (in Chinese)

PAPERNOT N , MCDANIEL P , JHA S , et al . The limitations of deep learning in adversarial settings [C]// 2016 IEEE European Symposium on Security and Privacy (EuroS&P) . Piscataway : IEEE , 2016 : 372 - 387 .

COMBEY T , LOISON A , FAUCHER M , et al . Probabilistic Jacobian-based saliency maps attacks [J]. Machine Learning and Knowledge Extraction , 2020 , 2 ( 4 ): 558 - 578 .

LIN Y , ZHAO H J , TU Y , et al . Threats of adversarial attacks in DNN-based modulation recognition [C]// IEEE INFOCOM 2020 - IEEE Conference on Computer Communications . Piscataway : IEEE , 2020 : 2469 - 2478 .

FLOWERS B , BUEHRER R M , HEADLEY W C . Evaluating adversarial evasion attacks in the context of wireless communications [J]. IEEE Transactions on Information Forensics and Security , 2020 , 15 : 1102 - 1113 .

ZHAO H J , LIN Y , GAO S , et al . Evaluating and improving adversarial attacks on DNN-based modulation recognition [C]// GLOBECOM 2020 - 2020 IEEE Global Communications Conference . Piscataway : IEEE , 2021 : 1 - 5 .

王超 , 魏祥麟 , 田青 , 等 . 基于特征梯度的调制识别深度网络对抗攻击方法 [J]. 计算机科学 , 2021 , 48 ( 7 ): 25 - 32 .

WANG C , WEI X L , TIAN Q , et al . Feature gradient-based adversarial attack on modulation recognition-oriented deep neural networks [J]. Computer Science , 2021 , 48 ( 7 ): 25 - 32 . (in Chinese)

HE K M , ZHANG X Y , REN S Q , et al . Deep residual learning for image recognition [C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2016 : 770 - 778 .

SADEGHI M , LARSSON E G . Adversarial attacks on deep-learning based radio signal classification [J]. IEEE Wireless Communications Letters , 2019 , 8 ( 1 ): 213 - 216 .

SAINATH T N , VINYALS O , SENIOR A , et al . Convolutional, long short-term memory, fully connected deep neural networks [C]// 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . Piscataway : IEEE , 2015 : 4580 - 4584 .

HOCHREITER S , SCHMIDHUBER J . Long short-term memory [J]. Neural Computation , 1997 , 9 ( 8 ): 1735 - 1780 .

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

一种量子概率启发的对话讽刺识别网络模型

面向时序异常检测的可变视距多向扫描方法

基于稀疏平滑自蒸馏的差分隐私深度学习方法

基于非一般类算子融合方法及硬件架构设计