北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2025, Vol. 48 ›› Issue (1): 33-38.

• 论文 • 上一篇    下一篇

基于双层注意力的异常数据深度分析及检测

郭高强1, 何明枢2, 李昕航3, 王小娟2   

  1. 1. 北京邮电大学 电子工程学院; 

    2. 北京邮电大学 网络空间安全学院; 3. 北京邮电大学 国际学院

  • 收稿日期:2023-12-05 修回日期:2024-01-27 出版日期:2025-02-26 发布日期:2025-02-25
  • 通讯作者: 何明枢 E-mail:hemingshu@bupt.edu.cn

Deep Analysis and Detection of Anomalous Data Based on Dual-Layer Attention

GUO Gaoqiang1, HE Mingshu2, LI Xinhang3, WANG Xiaojuan2   

  • Received:2023-12-05 Revised:2024-01-27 Online:2025-02-26 Published:2025-02-25
  • Contact: MINGSHU HE E-mail:hemingshu@bupt.edu.cn

摘要: 智能驾驶技术的发展促使车辆与外界网络联系日益频繁。控制器局域网络是当前主要车载网络,攻击者已经可以利用该网络中的安全漏洞实现车辆控制,从而对车内人员造成重大安全威胁。针对这一问题,提出了一个基于长短期记忆网络和注意力机制的异常检测方法,利用双层注意力编码器分别实现对控制器局域网络数据流局部和整体特征间信息的深度挖掘。通过高效搜寻并学习特征间的顺序模式,所提方法使用特征信息量明显减少,同时实现更好的检测效率和准确率,并能够实现不平衡的分类任务。最后,在CarHacking数据集上进行多次独立重复测试。结果表明,所提方法对数据集中所有攻击类别的检测正确率均高于99.20%,显著优于现有检测方法。此外,还完成了其他方法未实现的多分类任务,并且所提方法的平均检测正确率达到99.26%。

关键词: 车载网络, 异常检测, 长短期记忆模型, 注意力机制

Abstract: The development of intelligent driving technology increased vehicles interactions with external networks. The controller area network is the primary in-vehicle network and its security vulnerabilities can be exploited by attackers to gain control of vehicles, posing significant safety threats to occupants. To address this issue, an anomaly detection method based on long short-term memory and an attention mechanism is proposed. A dual-layer attention encoder is employed to deeply extract information from both local and global features within the data flow of the controller area network. By efficiently identifying and learning sequential patterns among features, the proposed method requires significantly less feature information, achieves better detection efficiency and accuracy, and enables unbalanced classification tasks. Finally, multiple independent repeated tests are performed on the CarHacking dataset. The results demonstrate that detection accuracy exceeding 99.20% has been achieved for all attack categories in the dataset, significantly outperforming existing detection methods. Additionally, a multi-classification task that could not be achieved by other approaches is accomplished by the proposed model, with an average detection accuracy of 99.26% .

Key words: vehicle network, anomaly detection, long short-term memory network, attention mechanism

中图分类号: