This python code simulates EEG in an eyes-closed scenario (patient has their eyes closed). There are 32 channel according to the standard 10-20 system, plus one additional channel that shows the amount of alpha waves in the recording.

The data is created for a sampling rate of 500 Hz, each channel contains an Ornstein-Uhlenbeck process (coloured noise), plus Alpha waves for some channels (sine at 10 Hz). The eeg intensity for Fp1 and Fp2 as well as for Cz is a bit higher than in the other channels. Alpha intensity for occipital channels (O1, O2) is high, for neighbouring channels significantly lower, central and frontal channels have no alpha.

The system is intended to simulate a realtime recording and export via LSL, to test EEG display and/or analysis. On my laptop (2GHz) 10 seconds of EEG data at 500 Hz sampling rate are created in 3 seconds.