一种基于集成学习的列车控制系统入侵检测方法  

An Intrusion Detection Method of Train Control System Based on Ensemble Learning

作　　者：王浩洋 李伟 彭思维 秦元庆[1,2] WANG Haoyang;LI Wei;PENG Siwei;QIN Yuanqing(School of Artificial Intelligence and Automation,Huazhong University of Science and Technology,Wuhan 430074,China;MOE Laboratory of Image Processing and Intelligent Control,Wuhan 430074,China;Intelligent Network R&D Department of CRRC Zhuzhou Locomotive Co.,Ltd.,Zhuzhou 412001,China;The State Key Laboratory of Heavy Duty AC Drive Electric Locomotive Systems Integration of CRRC Zhuzhou Locomotive Co.,Ltd.,Zhuzhou 412001,China)

机构地区：[1]华中科技大学人工智能与自动化学院,武汉430074 [2]华中科技大学图像信息处理与智能控制教育部重点实验室,武汉430074 [3]中车株洲电力机车有限公司智能网络研发部,株洲412001 [4]中车株洲大功率交流传动电力机车系统集成国家重点实验室,株洲412001

出　　处：《信息网络安全》2022年第5期46-53,共8页Netinfo Security

基　　金：国家自然科学基金[61873103]。

摘　　要：文章提出一种基于集成学习的列车控制系统入侵检测方法,采用随机森林分类器实现一维多尺度卷积网络空间特征提取模型与自适应时间卷积网络时间特征提取模型的融合,降低网络泛化误差,提高入侵检测准确率。基于列车控制系统半实物仿真平台模拟的入侵检测数据集,文章对该方法进行实验评估和对比测试,结果证明,该方法更具优势。This paper proposes an intrusion detection method based on ensemble learning.The random forest classifier is used to integrate the spatial feature extraction model of one-dimensional multi-scale convolution network and the temporal feature extraction model of adaptive time convolution network,so as to reduce the network generalization error and improve the accuracy of intrusion detection.Based on the intrusion detection data set simulated by the hardware-in-the-loop simulation platform of train control system,this paper conducts experimental evaluation and comparative tests on the proposed intrusion detection method,and the results prove the advantages of the method.

关 键 词：列车控制系统 集成学习 入侵检测 卷积神经网络 自适应计算时间 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]