Skip to main content

​
CAN Reverse Engineering

Automated payload reverse engineering pipeline for the Controller Area Network (CAN) protocol using machine learning

​
Overview

This project provides Python and R scripts for automated reverse engineering of Controller Area Network (CAN) payloads observed from passenger vehicles. Originally developed by Dr. Brent Stone at the Air Force Institute of Technology, this research tool uses machine learning techniques to analyze and decode proprietary CAN bus signals.

Get Started

Install and run your first CAN analysis in minutes

Pipeline Overview

Understand the three-phase reverse engineering process

API Reference

Explore the core classes and modules

Advanced Features

Multi-file batch processing and validation

​
Key Features

Lexical Analysis

Bit-level tokenization and signal extraction from CAN payloads

Semantic Analysis

Correlation-based clustering to identify related signals

J1979 Detection

Automatic identification of SAE J1979 standard signals

Time Series Visualization

Plot and analyze signal behavior over time

Multi-File Processing

Batch process multiple CAN log files automatically

EDM Integration

Empirical Dynamic Modeling with R scripts

​
Quick Example

Main.py
# Run the basic pipeline
python Main.py loggerProgram0.log

# Process can-utils format
python Main.py -c inputFile.log
The pipeline automatically:
  1. Pre-processes CAN data and identifies J1979 signals
  2. Tokenizes payloads to extract individual signals
  3. Clusters signals by correlation
  4. Generates visualizations and reports

​
Research Foundation

This code was developed in pursuit of a Ph.D. in Computer Science. For details about the methods and algorithms used, see the included dissertation: “Enabling Auditing and Intrusion Detection for Proprietary Controller Area Networks” (AFIT-END-DS-18-D-003.pdf).
The views expressed in this project are those of the author and do not reflect the official policy or position of the United States Air Force, Department of Defense, or United States Government.

Build docs developers (and LLMs) love

Get started for freeTalk to us