CAN-LOC: Spoofing Detection and Physical Intrusion Localization on an In-Vehicle CAN Bus Based on Deep Features of Voltage Signals

The Controller Area Network (CAN) is used for communication between in-vehicle devices. The CAN bus has been shown to be vulnerable to remote attacks. To harden vehicles against such attacks, vehicle manufacturers have divided invehicle networks into sub-networks, logically isolating critical devices. However, attackers may still have physical access to various sub-networks where they can connect a malicious device. This threat has not been adequately addressed, as methods proposed to determine physical intrusion points have shown weak results, emphasizing the need to develop more advanced techniques. To address this type of threat, we propose a security hardening system for in-vehicle networks. The proposed system includes two mechanisms that process deep features extracted from voltage signals measured on the CAN bus. The first mechanism uses data augmentation and deep learning to detect and locate physical intrusions when the vehicle starts; this mechanism can detect and locate intrusions, even when the connected malicious devices are silent. This mechanism’s effectiveness (100% accuracy) is demonstrated in a wide variety of insertion scenarios on a CAN bus prototype. The second mechanism is a continuous device authentication mechanism, which is also based on deep learning; this mechanism’s robustness (99.8% accuracy) is demonstrated on a real moving vehicle.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Christopher Huth,et al.  Scission: Signal Characteristic-Based Sender Identification and Intrusion Detection in Automotive Networks , 2018, CCS.

[3]  Azeem Hafeez,et al.  ECU Fingerprinting through Parametric Signal Modeling and Artificial Neural Networks for In-vehicle Security against Spoofing Attacks , 2019, 2019 15th International Computer Engineering Conference (ICENCO).

[4]  Pal-Stefan Murvay,et al.  TIDAL-CAN: Differential Timing Based Intrusion Detection and Localization for Controller Area Network , 2020, IEEE Access.

[5]  Amir Moradi,et al.  DL-LA: Deep Learning Leakage Assessment: A modern roadmap for SCA evaluations , 2019, IACR Cryptol. ePrint Arch..

[6]  Benjamin Timon,et al.  Non-Profiled Deep Learning-Based Side-Channel Attacks , 2019, IACR Cryptol. ePrint Arch..

[7]  Mathias Payer,et al.  Evading Voltage-Based Intrusion Detection on Automotive CAN , 2021, NDSS.

[8]  Ludovic Apvrille,et al.  Car2X Communication: Securing the Last Meter - A Cost-Effective Approach for Ensuring Trust in Car2X Applications Using In-Vehicle Symmetric Cryptography , 2011, 2011 IEEE Vehicular Technology Conference (VTC Fall).

[9]  Radha Poovendran,et al.  Cloaking the Clock: Emulating Clock Skew in Controller Area Networks , 2017, 2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS).

[10]  Longbing Cao,et al.  Training deep neural networks on imbalanced data sets , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[11]  Liang Huang,et al.  Data Augmentation for Deep Learning-Based Radio Modulation Classification , 2019, IEEE Access.

[12]  Shreyas Sen,et al.  X-DeepSCA: Cross-Device Deep Learning Side Channel Attack* , 2019, 2019 56th ACM/IEEE Design Automation Conference (DAC).

[13]  Christopher Huth,et al.  EASI: Edge-Based Sender Identification on Resource-Constrained Platforms for Automotive Networks , 2020, NDSS.

[14]  Stacy J. Prowell,et al.  Modeling inter-signal arrival times for accurate detection of CAN bus signal injection attacks: a data-driven approach to in-vehicle intrusion detection , 2017, CISRC.

[15]  Jiajia Liu,et al.  In-Vehicle Network Attacks and Countermeasures: Challenges and Future Directions , 2017, IEEE Network.

[16]  Kang G. Shin,et al.  Fingerprinting Electronic Control Units for Vehicle Intrusion Detection , 2016, USENIX Security Symposium.

[17]  Liam Kilmartin,et al.  Intra-Vehicle Networks: A Review , 2015, IEEE Transactions on Intelligent Transportation Systems.

[18]  Liang Xiao,et al.  Voltage Based Authentication for Controller Area Networks with Reinforcement Learning , 2019, ICC 2019 - 2019 IEEE International Conference on Communications (ICC).

[19]  Ming Li,et al.  SIMPLE: single-frame based physical layer identification for intrusion detection and prevention on in-vehicle networks , 2019, ACSAC.

[20]  Assaf Harel,et al.  Optimizing CAN Bus Security with In-Place Cryptography , 2019 .

[21]  Eric Sax,et al.  CAN Radar: Sensing Physical Devices in CAN Networks based on Time Domain Reflectometry , 2019, 2019 IEEE Vehicular Networking Conference (VNC).

[22]  Vyas Sekar,et al.  CANvas: Fast and Inexpensive Automotive Network Mapping , 2019, USENIX Security Symposium.

[23]  Kang G. Shin,et al.  Viden: Attacker Identification on In-Vehicle Networks , 2017, CCS.

[24]  Dong Hoon Lee,et al.  A Practical Wireless Attack on the Connected Car and Security Protocol for In-Vehicle CAN , 2015, IEEE Transactions on Intelligent Transportation Systems.

[25]  Mani Mina,et al.  Physical-Layer Identification of Wired Ethernet Devices , 2012, IEEE Transactions on Information Forensics and Security.

[26]  Bogdan Groza,et al.  Source Identification Using Signal Characteristics in Controller Area Networks , 2014, IEEE Signal Processing Letters.

[27]  Edison Pignaton de Freitas,et al.  A Quantitative Performance Study on CAN and CAN FD Vehicular Networks , 2018, IEEE Transactions on Industrial Electronics.

[28]  Stefano Zanero,et al.  A Stealth, Selective, Link-Layer Denial-of-Service Attack Against Automotive Networks , 2017, DIMVA.

[29]  Feng Duan,et al.  A Novel Deep Learning Approach With Data Augmentation to Classify Motor Imagery Signals , 2019, IEEE Access.

[30]  Christopher Huth,et al.  On the Robustness of Signal Characteristic-Based Sender Identification , 2019, ArXiv.

[31]  Dong Hoon Lee,et al.  A Practical Security Architecture for In-Vehicle CAN-FD , 2016, IEEE Transactions on Intelligent Transportation Systems.

[32]  Yun Yang,et al.  Identify a Spoofing Attack on an In-Vehicle CAN Bus Based on the Deep Features of an ECU Fingerprint Signal , 2020, Smart Cities.

[33]  Dong Hoon Lee,et al.  Identifying ECUs Using Inimitable Characteristics of Signals in Controller Area Networks , 2016, IEEE Transactions on Vehicular Technology.