Automotive cyber security
During the recent decades, the automotive industry has seen an increased amount of software in vehicles. This is a consequence of the addition of ADAS functions among others. Features that were once only seen in high-end vehicles can now be found almost everywhere. As vehicles become more automated, it can be expected that the importance of software applications will continue to grow. However, this change also comes with a potential risk, which is the security aspect.
Together with a partner, Volvo Group is developing an automated and intelligent layer of protection against cyberattacks, frauds and misuse by enabling detection, monitoring and response to attacks targeting any part of the connected vehicle framework.
As vehicles become more automated, it can be expected that the importance of software applications will continue to grow. However, this change also comes with a potential risk, which is the security aspect.
To ensure that the software remains functional and behaves as intended, safety measures are required in order to protect vehicles from cyberattacks.
Unsupervised multilayered models are used to describe the normal activity of a single vehicle, vehicle clusters, and the entire mobility service.
Unknown threats are detected by comparing anomalies from one or more of the machine learning models with a normal vehicle’s or vehicle cluster’s activity.
Furthermore, multiple anomalies are scored and aggregated to provide detection of unusual activity over time or across multiple vehicles, which helps to optimize the detection rate and reduce the rate of false positives.
The result of this process is early and accurate detection of actionable incidents that enable effective root cause analysis.
|
2023-01-13 01:47
File (PDF, Word, PPT, etc)
Connected vehicles - The more things are connected, the higher the security concern. |