Multi-Sensor Device for Police Firearm Monitoring Raw Dataset

The following dataset consists of pre-labeled multivariate time-series vectors collected from a firearm monitoring device. It includes data from a piezoelectric sensor, which captures vibrations generated by the firearm, and a photoelectric sensor, which measures the remaining number of rounds in the firearm's magazine. The labeled classes in the dataset represent typical firearm manipulations, including shots, reloads, and other related actions.
The data was collected between March and May 2025 at the shooting training camp in Lleida by police instructors and specialized officers. The structure and format of the dataset are designed to support the training of deep learning algorithms for classifying event labels or extracting other metadata from the combined sensor data.