1.School of Cyber Science and Engineering,Wuhan University,Wuhan 430072,Hubei,China
2.Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ),Shenzhen 518123,Guangdong,China
Show less
文章历史+
Received
Published
2023-05-30
2023-10-24
Issue Date
2026-07-23
PDF (1170K)
摘要
为解决车联网中CAN(controller area network)总线易受攻击的问题,提出了一个混合攻击入侵检测模型DGAO-IDS。该模型利用无监督的DBN(deep belief nets)学习正常CAN报文数据的基础特征,并利用一个带注意力机制的GRU(gate recurrent unit)网络学习其时序特征,用单分类支持向量机对其进行分类;引入一个对电子控制单元规则学习得到的过滤器,综合过滤器与前述模型的分类结果得出最终的检测结果。实验结果表明,对于不同攻击,基于规则的过滤器的假阳率均为0;DGAO-IDS模型不仅在公开数据集HCRL中的检测结果优于对比模型HyDL-IDS和MD-LSTM,而且该模型在混合攻击数据集MixAt中的精确度达到了91.05%,与HyDL-IDS模型和MD-LSTM模型相比分别高6.55%与7.93%。
Abstract
To address the vulnerability of the CAN(controller area network) bus in Telematics, this paper proposes a hybrid attack intrusion detection model, DGAO-IDS, which uses an unsupervised DBN to learn the basic features of normal CAN message data and a GRU network with an attention mechanism to learn its temporal features, and finally classifies them using a single classification support vector machine. The final detection result is obtained by combining the filter with the classification results of the aforementioned model. Experimental results demonstrate that the proposed rule-based filters have a false positive rate of zero for different attacks. Moreover, the DGAO-IDS model outperforms the comparative models HyDL-IDS and MD-LSTM on the public dataset HCRL. In addition, the model achieves an accuracy of 91.05% on the mixed-attack dataset MixAt, which is 6.55% and 7.93% higher than the HyDL-IDS model and the MD-LSTM model, respectively.
随着新能源汽车的高速发展,智能网联汽车(intelligent connected vehicle,ICV)已成为汽车行业发展的主要趋势。ICV将车联网与智能汽车有机联合,是配备了先进的智能传感器、执行器、电子控制单元(electronic control unit,ECU),与互联网相连,并拥有复杂环境感知和智能控制能力的智能汽车[1]。在智能汽车中,每个ECU在控制系统中完成一个特定的功能,如油门、刹车和转向的控制。ECU根据其功能或通信速率被划分为不同的子网进行通信。车载网络(vehicle-mounted network,VMN)通过网关将多个子网进行互联。一般来说,ECU通过控制器局域网(controller area network,CAN)进行通信。CAN与局域互联网(local interconnect network,LIN)以及FlexRay[2]共同组成了最常见的车载网络。与常见的以太网网络不同,CAN总线广播了一些包含车辆状态信息的短消息,以维持系统中所有ECU的数据一致性[3]。
MixAt数据集是在CAN总线仿真模拟环境下收集的数据。为了检验DGAO-IDS模型的功能,MixAt数据集只包含一个数据包,其数据类型如表2所示,该数据包记录了一个长达60 min的CAN总线运行(仿真环境)的数据。其中前35 min CAN总线是正常运行的,而后在CAN总线上找一个ECU充当受攻击节点,通过该ECU向正常运行的CAN总线中随机地发送一些恶意的攻击报文(例如DoS攻击报文和Fuzzy攻击报文)达到模拟混合攻击的效果。攻击报文是参考文献[23,27,35]的思想进行构造的。
ZHOUH B, XUW C, CHENJ C, et al. Evolutionary V2X technologies toward the Internet of vehicles: Challenges and opportunities[J]. Proceedings of the IEEE, 2020, 108(2): 308-323. DOI: 10.1109/JPROC.2019.2961937 .
[2]
LEET Y, KUOC W, LINI A. High performance CAN/ FlexRay gateway design for in-vehicle network[C]//2017 IEEE Conference on Dependable and Secure Computing. New York: IEEE Press, 2017: 240-242. DOI: 10.1109/DESEC.2017.8073848 .
[3]
LOKMANS F, OTHMANA T, ABU-BAKARM H. Intrusion detection system for automotive Controller Area Network (CAN) bus system: A review[J]. EURASIP Journal on Wireless Communications and Networking, 2019, 2019(1): 1-17. DOI: 10.1186/s13638-019-1484-3 .
[4]
MILLERC, VALASEKC. Remote Exploitation of An Unaltered Passenger Vehicle[DB/OL]. [2022-12-12]. DOI: 10.4324/9781315145891-7 .
[5]
NIES, LIUL, DUY. Free-fall: Hacking Tesla from Wireless to Can Bus[DB/OL]. [2022-12-16].
[6]
GREENBERGA. Hackers Can Steal a Tesla Model S in Seconds by Cloning Its Key Fob [DB/OL]. [2018-09-19].
[7]
WANGQ Y, SAWHNEYS. VeCure: A practical security framework to protect the CAN bus of vehicles[C]//2014 International Conference on the Internet of Things (IoT). New York: IEEE Press, 2015: 13-18. DOI: 10.1109/IOT.2014.7030108 .
[8]
LO W, ALQAHTANIH, THAKURK, et al. A hybrid deep learning based intrusion detection system using spatial-temporal representation of in-vehicle network traffic[J]. Vehicular Communications, 2022, 35: 100471. DOI: 10.1016/j.vehcom.2022.100471 .
[9]
SUNH, CHENM M, WENGJ, et al. Anomaly detection for in-vehicle network using CNN-LSTM with attention mechanism[J]. IEEE Transactions on Vehicular Technology, 2021, 70(10): 10880-10893. DOI: 10.1109/TVT.2021.3106940 .
[10]
DESTAA K, OHIRAS, ARAII, et al. U-CAN: A convolutional neural network based intrusion detection for controller area networks[C]//2022 IEEE 46th Annual Computers, Software, and Applications Conference (COMPSAC). New York: IEEE Press, 2022: 1481-1488. DOI: 10.1109/COMPSAC54236.2022.00235 .
[11]
QINH M, YANM R, JIH J. Application of Controller Area Network (CAN) bus anomaly detection based on time series prediction[J]. Vehicular Communications, 2021, 27: 100291. DOI: 10.1016/j.vehcom.2020.100291 .
[12]
WANGQ, LUZ J, QUG. An entropy analysis based intrusion detection system for controller area network in vehicles[C]//2018 31st IEEE International System-on-Chip Conference (SOCC). New York: IEEE Press, 2019: 90-95. DOI: 10.1109/SOCC.2018.8618564 .
[13]
JAVEDA R, REHMANS U, KHANM U, et al. CANintelliIDS: Detecting in-vehicle intrusion attacks on a controller area network using CNN and attention-based GRU[J]. IEEE Transactions on Network Science and Engineering, 2021, 8(2): 1456-1466. DOI: 10.1109/TNSE.2021.3059881 .
[14]
HUR, WUZ Y, XUY, et al. Multi-attack and multi-classification intrusion detection for vehicle-mounted networks based on mosaic-coded convolutional neural network[J]. Scientific Reports, 2022, 12: 6295. DOI: 10.1038/s41598-022-10200-4 .
[15]
SCHÖLKOPFB, WilliamsonR, SMOLAA,et al. Support vector method for novelty detection [DB/OL] .[2022-12-30] .DOI: 10.1162/089976600300015565 .
[16]
OGUMAH, YOSHIOKAA, NISHIKAWAM, et al. New attestation based security architecture for in-vehicle communication[C]//IEEE GLOBECOM 2008 ― 2008 IEEE Global Telecommunications Conference. New York: IEEE Press, 2008: 1-6. DOI: 10.1109/GLOCOM.2008.ECP.369 .
[17]
VAN HERREWEGEV, SINGELEED, VERBAUWHEDEI. CANAuth ― A simple, backward compatible broadcast authentication protocol for CAN bus[DB/OL]. [2011-12-02].
[18]
LINC W, SANGIOVANNI-VINCENTELLIA. Cyber-security for the controller area network (CAN) communication protocol[C]//2012 International Conference on Cyber Security. New York: IEEE Press, 2013: 1-7. DOI: 10.1109/CyberSecurity.2012.7 .
[19]
GROZAB, MURVAYS. Efficient protocols for secure broadcast in controller area networks[J]. IEEE Transactions on Industrial Informatics, 2013, 9(4): 2034-2042. DOI: 10.1109/TII.2013.2239301 .
[20]
SZILAGYIC, KOOPMANP. Low cost multicast authentication via validity voting in time-triggered embedded control networks[C]//Proceedings of the 5th Workshop on Embedded Systems Security. New York: ACM, 2010: 1-10. DOI: 10.1145/1873548.1873558 .
[21]
UEDAH, KURACHIR, TAKADAH, et al. Security authentication system for in-vehicle network[J]. SEI technical review, 2015, 81: 5-9.
[22]
FARAGW A. CANTrack: Enhancing automotive CAN bus security using intuitive encryption algorithms[C]//2017 7th International Conference on Modeling, Simulation, and Applied Optimization (ICMSAO). New York: IEEE Press, 2017: 1-5. DOI: 10.1109/ICMSAO.2017.7934878 .
[23]
OTHMANE LBEN, DHULIPALAL, ABDELKHALEKM, et al. On the performance of detecting injection of fabricated messages into the CAN bus[J]. IEEE Transactions on Dependable and Secure Computing, 2022, 19(1): 468-481. DOI: 10.1109/TDSC.2020.2990192 .
[24]
OLUFOWOBIH, YOUNGC, ZAMBRENOJ, et al. SAIDuCANT: Specification-based automotive intrusion detection using controller area network (CAN) timing[J]. IEEE Transactions on Vehicular Technology, 2020, 69(2): 1484-1494. DOI: 10.1109/TVT.2019.2961344 .
[25]
TAYLORA, JAPKOWICZN, LEBLANCS. Frequency-based anomaly detection for the automotive CAN bus[C]//2015 World Congress on Industrial Control Systems Security (WCICSS). New York: IEEE Press, 2016: 45-49. DOI: 10.1109/wcicss.2015.7420322 .
[26]
ZHUK L, CHENZ C, PENGY Y, et al. Mobile edge assisted literal multi-dimensional anomaly detection of in-vehicle network using LSTM[J]. IEEE Transactions on Vehicular Technology, 2019, 68(5): 4275-4284. DOI: 10.1109/TVT.2019.2907269 .
[27]
LIX H, ZHANGH Y, MIAOY B, et al. CAN bus messages abnormal detection using improved SVDD in Internet of vehicles[J]. IEEE Internet of Things Journal, 2022, 9(5): 3359-3371. DOI: 10.1109/JIOT.2021.3098221 .
[28]
CHENGP Z, XUK, LIS M, et al. TCAN-IDS: Intrusion detection system for Internet of vehicle using temporal convolutional attention network[J]. Symmetry, 2022, 14(2): 310. DOI: 10.3390/sym14020310 .
[29]
JEDHM, OTHMANE LBEN, AHMEDN, et al. Detection of message injection attacks onto the CAN bus using similarities of successive messages-sequence graphs[J]. IEEE Transactions on Information Forensics and Security, 2021, 16: 4133-4146. DOI: 10.1109/TIFS.2021.3098162 .
[30]
ISLAMR, DEVNATHM K, SAMADM D, et al. GGNB: Graph-based Gaussian naive Bayes intrusion detection system for CAN bus[J]. Vehicular Communications, 2022, 33: 100442. DOI: 10.1016/j.vehcom.2021.100442 .
[31]
REFATR U D, ELKHAILA A, MALIKH. A lightweight intrusion detection system for CAN protocol using neighborhood similarity[C]//2022 7th International Conference on Data Science and Machine Learning Applications (CDMA). New York: IEEE Press, 2022: 121-126. DOI: 10.1109/CDMA54072.2022.00025 .
MAW G, ZHANGY D, GUOJ. Abnormal traffic detection method based on LSTM and improved residual neural network optimization[J]. Journal on Communications, 2021, 42(5): 23-40. DOI: 10.11959/j.issn.1000-436x.2021109(Ch ).
[36]
LEEH, JEONGS H, KIMH K. OTIDS: A novel intrusion detection system for in-vehicle network by using remote frame[C]//2017 15th Annual Conference on Privacy, Security and Trust (PST). New York: IEEE Press, 2018: 57-66. DOI: 10.1109/PST.2017.00017 .
[37]
KANGH, KWAKB, LEEY H, et al. Car Hacking: Attack and Defense Challenge 2020 Dataset[DB/OL]. [2022-02-03].