CSTE: A Dataset for Cross-Sessions and Cross-Tasks EEG Biometrics

This dataset is a self-collected identity recognition dataset specifically designed for cross-session and cross-task identity recognition, a task for which there are few existing datasets. According to the international 10-20 electrode system, EEG signals were recorded from 10 subjects (aged 20\u201325 years, without neurological disorders), all of whom signed informed consent forms. The EEG acquisition device used was the 64-channel wet electrode EEG amplifier ERP System ActiCHamp from Brain Products, with a sampling frequency of 500 Hz. The experiment consisted of two sessions, with an interval of approximately one week to one month to ensure the uniqueness of EEG distribution across different sessions. In each session, the subjects were asked to complete four dynamic or resting-state tasks, including mental arithmetic, motor imagery, resting, and passive audio. A rest period was set between each task.