智能网联汽车车载入侵检测系统测试数据库构建方法  

Construction Method of the Test Database of On-board Intrusion Detection System for Intelligent Vehicle

作　　者：罗勇 白日光 LUO Yong;BAI Riguang

机构地区：[1]联合汽车电子有限公司,上海201206

出　　处：《上海汽车》2021年第11期14-19,共6页Shanghai Auto

摘　　要：车载入侵检测系统(IDS)是一项加强车内网络信息安全防护能力的措施,然而目前业界针对车载入侵检测系统的研究没有统一的测试数据库作为基准。文章提出了一种测试数据库的构建方法,构建了一套针对车内CAN总线网络的入侵检测系统测试数据库,来解决业界研究所用测试数据不同且评判标准不统一的问题。首先,文章对车载入侵检测系统的研究现状进行了调研总结;其次,基于车内网络面临的常见安全威胁进行了概括,主要包括拒绝服务、信号篡改、伪造信号、报文注入、信号重放等攻击模式,再根据车内网络面临的安全威胁提出了测试数据库的构建方法;最后,利用LSTM(Long Short Term Memory)算法设计了一套入侵检测算法,来验证测试数据库的有效性。结果表明,该数据库可用于车载入侵检测算法的验证。On-board intrusion detection system(IDS)is a countermeasure to strengthen the security of in-vehicle network.However,there is no unified test database for the research of this area in the industry as a standard.This paper proposes a method for constructing a test database for the on-board IDS system focusing the CAN bus network in order to solve the problem of lacking a unified standard.Firstly,the research situation of on-board IDS is summarized.Secondly,common security threats faced by in-vehicle networks are summarized,including denial of service,signal tampering,signal spoofing,message injection,signal replay,etc.Based on the summarized security threats,a method is proposed for the construction of test database.Finally,an intrusion detection algorithm is designed using LSTM(Long Short Term Memory)algorithm,to verify the validity of the test database.The results show that the database can be used in the verification of on-board IDS.

关 键 词：智能网联汽车 信息安全 入侵检测 数据库 

分 类 号：TN915.08[电子电信—通信与信息系统] U463.6[电子电信—信息与通信工程]