Reverse Engineering 17+ Cars in Less Than 10 Minutes

The presentation discusses the process of slicing and dicing data to identify different types of information in vehicles and medical devices for cybersecurity purposes.

• The process involves lexical and semantic analysis to identify distinct pieces of information
• The example given involves identifying vehicle speed and RPM per wheel from 64-bit payloads
• The process can be automated to teach a computer to identify different types of information
• The presentation provides a demo of the process and offers resources on Github for further exploration

The presenter used an example of identifying vehicle speed and RPM per wheel from a car's 64-bit payloads to illustrate the process of slicing and dicing data. The computer was able to identify these distinct pieces of information automatically, which can be useful for cybersecurity purposes.

Brent provides a live demonstration reversing engineering 17 or more unknown passenger vehicle CAN networks in under 10 minutes using new automated techniques. These unsupervised techniques are over 90% accurate and consistent when tested using production CAN networks and different driving conditions. He then introduces the Python and R code used for the demo and posted to his public GitHub repository at https://github.com/brent-stone/CAN_Reverse_Engineering. The Dissertation explaining how the code works is also posted.