.png){kind=icon}
返回
MTDCAP: Moving Target Defense-Based CAN Authentication Protocol
.png){kind=icon}
PRE
AI
DOI：10.1109/TITS.2024.3384054.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The convenience behind modern intelligent vehicles is simply that a group of intelligent electronic control units (ECUs) connected to controller area network (CAN) work in concert. However, quite a lot of studies have shown their security concerns about CAN. In this era of rampant cyberattacks, due to the broadcast mechanism of CAN and the lack of necessary security mechanisms such as encryption and authentication, ECUs are easily disturbed by various cyberattacks, thus leading to vehicle failures. To solve this problem, we propose a novel security authentication protocol based on the core concept of moving target defense, namely MTDCAP, which utilizes MaskedID and hash chains to maintain anonymity externally and ensure the authentication of the sender internally. Unlike general hash chain-based authentication, our protocol creatively incorporates a self-renewal mechanism into the hash chain, which effectively reduces the time overhead and security risk of negotiating the update of the hash chain between both communicating parties. In addition, AES serves to encrypt CAN message payload so as to prevent adversaries from eavesdropping. The theoretical analysis for the security against four kinds of attacks (i.e., eavesdropping, impersonation, replay, and bus-off attacks) in MTDCAP is detailed. Afterwards, a series of protocol evaluations are conducted on two kinds of typical hardware platforms, including T-Box from real vehicle supported by XPeng, and the results reveal that the proposed protocol significantly outperforms the existing protocols in the robustness, bus load, and time overhead. In particular, the authentication overhead on T-Box is only 0.18 ms for MTDCAP.
Keyword:
Protocols
Authentication
Feature extraction
Encryption
Cryptography
Computer crime
Sun
Automotive security
controller area network (CAN)
CAN bus
moving target defense
authentication protocol
期刊
IF:
8.4
论文数:
9.6K
被引数:
6.3W
机构
引用论文
CAN-LOC: Spoofing Detection and Physical Intrusion Localization on an In-Vehicle CAN Bus Based on Deep Features of Voltage SignalsCAN-LOC: 基于电压信号深层特征的车载CAN总线欺骗检测与物理入侵定位
OAAI
.png){kind=icon}
OAAI分享收藏分享收藏Detecting Vehicle Anomaly in the Edge via Sensor Consistency and Frequency Characteristic基于传感器一致性和频率特性的边缘车辆异常检测
PREAI
分享收藏分享收藏
分享收藏
查看更多内容