<![CDATA[The development of modern automotive technology emphasizes the importance of vehicle connectivity and autonomy, with the aim of enhancing safety and comfort. Due to its role in managing critical vehicle functions and its vulnerability to security threats, the automotive industry has developed two primary approaches to address CAN Bus security. Therefore, the automotive industry has developed two main approaches to address this security issue. First, passive defense through security protocols that include encryption, authentication, and message verification, and second, anomaly detection using advanced technologies. Several anomaly detection methods have been introduced, including K-Means clustering, Support Vector Machines, and Deep Learning, each offering advantages in detecting specific attack patterns. However, one increasingly popular approach is time analysis, which leverages message inter-arrival patterns and clock skew on the CAN Bus to identify suspicious behavior and detect anomalies in real-time. Although this method has shown effectiveness in detecting various types of attacks, the main challenge lies in its ability to identify highly concealed attacks that may not be visible to traditional methods. This research provides an understanding of various anomaly detection approaches on electric vehicle CAN Bus networks, with a primary focus on time analysis. By reviewing several approaches, this study offers valuable insights into improving the security of electric vehicles and the overall smart transportation ecosystem. The results show that time-based analysis methods can detect various types of attacks, such as spoofing, replay attacks, and denial-of-service, with high efficiency.]]>
Copyrights © 2025
