信息网络安全 ›› 2026, Vol. 26 ›› Issue (8): 1264-1276.doi: 10.3969/j.issn.1671-1122.2026.08.008

• 学术研究 • 上一篇    下一篇

Argus6:一种IPv6 DDoS威胁检测模型

张宏涛1,2, 王清涛1,2, 张连成3(), 王吉昌3   

  1. 1 郑州大学网络空间安全学院郑州 450002
    2 郑州大学网络管理中心郑州 450001
    3 网络空间部队信息工程大学网络空间安全学院郑州 450001
  • 收稿日期:2025-12-25 出版日期:2026-08-10 发布日期:2026-09-23
  • 通讯作者: 张连成 E-mail:liancheng17@aliyun.com
  • 作者简介:张宏涛(1977—),男,陕西,高级实验师,博士,主要研究方向为网络与系统安全、数据安全、下一代互联网安全|王清涛(2000—),男,河南,硕士研究生,主要研究方向为下一代互联网安全、IPv6网络安全|张连成(1982—),男,河南,副教授,博士,主要研究方向为下一代互联网安全、IPv6网络安全、SDN网络安全|王吉昌(1999—),男,河南,助教,硕士,主要研究方向为下一代互联网安全、IPv6网络安全
  • 基金资助:
    国家自然科学基金(61402526)

Argus6: an IPv6 DDoS threat detection model

Zhang Hongtao1,2, Wang Qingtao1,2, Zhang Liancheng3(), Wang Jichang3   

  1. 1 School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450002, China
    2 Network Management Center, Zhengzhou University, Zhengzhou 450001, China
    3 School of Cyberspace Security, Cyberspace Force Information Engineering University, Zhengzhou 450001, China
  • Received:2025-12-25 Online:2026-08-10 Published:2026-09-23
  • Contact: Zhang Liancheng E-mail:liancheng17@aliyun.com

摘要:

IPv6网络的快速扩展加剧了分布式拒绝服务(DDoS)威胁的多样性和复杂性,而实际运行环境中标注流量数据的匮乏又严重削弱了传统监督式检测器的性能。为应对这些挑战,文章提出Argus6混合深度学习模型,该模型并行融合了基于Transformer的分支与Mamba状态空间分支,Transformer分支用于建模全局时序依赖关系,Mamba分支用于高效捕捉长序列动态特征。在两个基准数据集上的消融实验表明,完整的Argus6架构显著优于单一结构变体:在数据集1上,F1-Score为0.9874、AUC为0.9986;在数据集2上,F1-Score为0.9926、AUC为0.9995,且误报率及漏报率分别降至1.19%和0.84%。在K-shot小样本实验中,Argus6仅需少量标签即可快速收敛:每类仅使用10个样本时,F1-Score约为0.78、AUC约为0.81;当每类样本超过100个时,F1-Score约为0.91、AUC约为0.96。最后,在完整的IPv6 DDoS威胁数据集上,Argus6在准确率、F1-ScoreAUC和错误率等指标上均超越了1D-CNN、CNN-LSTM、MF-Net及多种经典方法,充分验证了其在大规模DDoS威胁防御中的鲁棒性与数据效率。

关键词: IPv6, DDoS威胁, transformer, mamba

Abstract:

This paper presented a hybrid deep-learning method, Argus6, for IPv6 DDoS threat detection under limited labeled data. The method parallelized a transformer-based branch to capture global temporal dependencies and a Mamba state-space branch to model long-sequence dynamics efficiently. Ablation studied on two benchmark datasets showed that the proposed method significantly outperformed its single-branch variants. On Dataset 1, it achieved an F1-Score of 0.9874 and an AUC of 0.9986; on Dataset 2, it obtained an F1-Score of 0.9926 and an AUC of 0.9995, with false positive and false negative rates reduced to 1.19% and 0.84%, respectively. In K-shot experiments, the method demonstrated strong data efficiency: with only 10 labeled samples per class, it yielded an F1-Score of approximately 0.78 and an AUC of approximately 0.81; when the number of samples per class exceeded 100, the F1-Score and AUC reached approximately 0.91 and 0.96, respectively. On the full IPv6 DDoS threat dataset, the method surpasses 1D-CNN, CNN-LSTM, MF-Net, and classical baselines in accuracy, F1-Score, AUC, and error-rate metrics, which confirms its robustness and data efficiency for large-scale DDoS threat defense.

Key words: IPv6, DDoS threat, transformer, mamba

中图分类号: