摘要
随着智能网联汽车技术的发展,其信息安全问题也日益凸显,面向智能网联汽车CAN网络入侵检测的需求,针对周期性攻击,提出了一种基于ID熵的检测模型,判断CAN网络通信的异常状态并对攻击ID进行定位;针对非周期性攻击,提出了一种基于支持向量机-数据关联性的入侵检测方案,并设计了车载CAN网络入侵检测装置。实验结果表明,对周期性攻击如重放攻击、DoS攻击和丢弃攻击的检测准确率为100%,对非周期性攻击如篡改攻击的检测准确率为97.14%。该装置可有效地检测CAN网络入侵行为,为智能网联环境下的整车信息安全提供有力保障。
With the development of intelligent connected vehicle technology,information security is becoming more and more important.For the requirements of intelligent networked vehicle intrusion detection,two detection models are proposed.For the periodic attacks,a detection scheme based on ID entropy is proposed,which can analysis and judge the abnormal state of CAN communication,in addition,the abnormal message can be located.For the aperiodic attacks,a detection scheme based on support vector machine-data relevance is proposed,and a vehicle-mounted CAN intrusion detection device is designed and developed.The experimental results show that the detection accuracy rate of periodic attacks is 100%,such as replay,DoS and discard attack,and the detection accuracy rate of aperiodic attacks is 97.14%,such as tampering attack.CAN intrusion can be effectively detected by the device,and the vehicle information security under the intelligent connected environment can be guaranteed.
作者
刘蓬勃
彭海德
赵剑
贾博文
范科峰
LIU Pengbo;PENG Haide;ZHAO Jian;JIA Bowen;FAN Kefeng(School of Automotive Engineering,Dalian University of Technology,Dalian 16024,China;China Electronics Standardization Institute,Beijing 100007,China)
出处
《实验技术与管理》
CAS
北大核心
2022年第3期126-131,170,共7页
Experimental Technology and Management
基金
国家重点研发计划(2018YFA0703200)
2019年工业和信息化领域公共服务能力提升专项(2019-00910-5-1)。
关键词
汽车控制器局域网络
入侵检测
ID熵
支持向量机
数据关联性
automotive controller area network(CAN)
intrusion detection
ID entropy
support vector machine
data relevance
作者简介
刘蓬勃(1981—),男,辽宁大连,博士,工程师,主要研究方向为智能网联汽车和传感器,pengboliu@dlut.edu.cn;通信作者:赵剑(1980—),男,河北石家庄,博士,教授,主要研究方向为汽车传感器和智能网联汽车,jzhao@dlut.edu.cn。
