Software uses deep learning to protect cars against cyber attacks

Software uses deep learning to protect cars against cyber attacks

New Products |
By Christoph Hammerschmidt

vXRay can be integrated into customers’ connected vehicles’ Security Operation Centers (SOC) independently of vehicle architecture or ECU sourcing. It can help customers uncover zero-day vulnerabilities, provide early detection of vehicle malfunctions and flag misuse and abuse problems.

The new system uses unsupervised machine learning paradigms in a fully autonomous process to establish the normal behavior of the vehicle without dependencies or previous knowledge of ECU properties and protocols. Once the behavioral baseline is established, the machine learning models can detect, categorize and flag any abnormal behavior and report it to the connected vehicles’ SOC for further analysis.

Existing solutions for anomaly detection rely on in-vehicle analysis of signatures and error codes based on partial data and using limited computational power, explains Hilik Stein, CTO and Co-founder of SafeRide. “OEMs are receiving limited and filtered information about their fleets resulting in suboptimal operational performance and exposure to zero-day cybersecurity attacks and safety risks,” he says. “Our technology enhances the effectiveness of the connected vehicles’ SOC by integrating the ability to analyze raw vehicle data and significantly amplify its detection power.”

Today, security measures installed in vehicles can only detect and prevent known attacks. “With ever-growing cybersecurity threats, OEMs need to assume that security breaches are inevitable and install intrusion detection capabilities that can detect unknown threats and facilitate remediation,” said Gil Reiter, Vice President of Product Management and Marketing at SafeRide. According to Reiter, vXRay provides OEMs and fleet operators with immediate intrusion detection capabilities with minimal or no modifications to the vehicle platform.


The vXRay technology was proven to effectively detect all cyberattacks and vehicle malfunctions in multiple vehicle models in customer testing, and is being implemented by several major automotive vendors as part of their 2019 security strategies.

More information:

Related articles:

Cloud-based system counters attacks against the CAN bus



If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles