《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (3): 864-871.DOI: 10.11772/j.issn.1001-9081.2024030327

• 网络空间安全 • 上一篇    下一篇

面向加密恶意流量检测模型的堆叠集成对抗防御方法

陈瑞龙1, 胡涛1(), 卜佑军1, 伊鹏1, 胡先君2, 乔伟2   

  1. 1.信息工程大学 信息技术研究所,郑州 450002
    2.网络通信与安全紫金山实验室,南京 211111
  • 收稿日期:2024-03-27 修回日期:2024-06-04 接受日期:2024-06-11 发布日期:2024-07-16 出版日期:2025-03-10
  • 通讯作者: 胡涛
  • 作者简介:陈瑞龙(2000—),男,河南鹤壁人,硕士研究生,主要研究方向:网络入侵检测、深度学习
    卜佑军(1978—),男,河南焦作人,副研究员,博士,主要研究方向:网络安全、拟态防御
    伊鹏(1977—),男,湖北黄冈人,研究员,博士,主要研究方向:网络空间安全、网络体系结构
    胡先君(1989—),男,安徽无为人,工程师,博士,主要研究方向:先进网络防御
    乔伟(1989—),男,山东烟台人,工程师,硕士,主要研究方向:先进网络防御。
  • 基金资助:
    国家自然科学基金资助项目(62176264)

Stacking ensemble adversarial defense method for encrypted malicious traffic detection model

Ruilong CHEN1, Tao HU1(), Youjun BU1, Peng YI1, Xianjun HU2, Wei QIAO2   

  1. 1.Information Technology Research Institute,Information Engineering University,Zhengzhou Henan 450002,China
    2.Purple Mountain Laboratories,Nanjing Jiangsu 211111,China
  • Received:2024-03-27 Revised:2024-06-04 Accepted:2024-06-11 Online:2024-07-16 Published:2025-03-10
  • Contact: Tao HU
  • About author:CHEN Ruilong, born in 2000, M. S. candidate. His research interests include network intrusion detection, deep learning.
    BU Youjun, born in 1978, Ph. D., associate research fellow. His research interests include network security, mimic defense.
    YI Peng, born in 1977, Ph. D., research fellow. His research interests include cyberspace security, network architecture.
    HU Xianjun, born in 1989, Ph. D., engineer. His research interests include advanced network defense.
    QIAO Wei, born in 1989, M. S., engineer. His research interests include advanced network defense.
  • Supported by:
    National Natural Science Foundation of China(62176264)

摘要:

当前,基于深度学习的流量分类模型已广泛应用于加密恶意流量分类,然而深度学习模型所面临的对抗样本攻击问题严重影响了这些模型的检测精度和可用性。因此,提出一种面向加密恶意流量检测模型的堆叠集成对抗防御方法D-SE(Detector-Stacking Ensemble)。D-SE采用堆叠集成学习框架,分为对抗防御层和决策层。对抗防御层用于检测潜在的对抗攻击流量样本,在该层中包括由残差网络(ResNet)、CNN-LSTM、ViT(Vision Transformer)这3种分类器以及多层感知机组成的对抗攻击检测器,多层感知机根据分类器预测概率的分布检测是否发生对抗攻击。为提高检测器的对抗样本检测效果,对检测器进行对抗训练。在决策层中设计一种基于投票和权重机制的联合决策模块,并通过择多判决机制和高权重者优先机制避免最终预测结果过度依赖部分分类器。在USTC-TFC2016数据集上对D-SE进行测试的结果表明:在非对抗环境下,D-SE的准确率达到96%以上;在白盒攻击环境下,D-SE的准确率达到89%以上。可见,D-SE具有一定的对抗防御能力。

关键词: 恶意流量分类, 深度学习, 对抗攻击, 防御机制, 堆叠集成学习框架

Abstract:

Currently, deep learning-based traffic classification models are used widely for encrypted malicious traffic classification. However, adversarial attack samples faced by deep learning models severely impact the detection accuracy and availability of these models. Therefore, an adversarial defense method for encrypted malicious traffic detection models was proposed, namely D-SE (Detector-Stacking Ensemble). D-SE employed a stacking ensemble learning framework, which was divided into an adversarial defense layer and a decision layer. The former was used to detect potential adversarial traffic samples, including three classifiers — Residual Network (ResNet), CNN-LSTM, and Vision Transformer (ViT), and a multilayer perceptron as an adversarial attack detector. Based on the predicted probability distribution of the classifiers, the existence of adversarial attack was detected by the multilayer perceptron. To improve the detection performance of the detector for adversarial samples, the detector was enhanced via adversarial training. In the decision layer, a joint decision module based on voting and weight mechanism was designed, and through a majority rule decision mechanism and a high-weight-preference mechanism, excessive dependence on some classifiers was alleviated in the final prediction. The performance of D-SE was tested on USTC-TFC2016 dataset, and the results show that the accuracy of D-SE is over 96% in the non-adversarial environment, and the accuracy of D-SE is more than 89% in the white-box attack environment. It can be seen that D-SE has certain ability of adversarial defense.

Key words: malicious traffic classification, deep learning, adversarial attack, defense mechanism, stacking ensemble learning framework

中图分类号: