CAN Signal Extraction and Translation Dataset

This dataset contains 40 Controller Area Network (CAN) traffic logs collected from a production vehicle in a controlled driving environment. The logs were captured while periodically sending OBD-II diagnostic queries, enabling cross-analysis between raw CAN traffic and diagnostic response data. Each log file corresponds to a specific PID queried during data collection. The dataset is designed to support research in reverse engineering of CAN specifications, signal extraction, semantic translation of payload data, and vehicle-network security analysis. All CAN frames include timestamps, CAN IDs, DLC values, and 8-byte payload fields. The dataset was originally used in research on automated discovery of CAN signal mappings and is released to facilitate further work in semantic feature extraction and anomaly detection for in-vehicle networks.