In the first stage of the project we used Matlab to test our analog circuit for EEG signals.
We used a commercial National Instruments DAQ (NI usb-6009) to sample the output of the analog circuit.
The Matlab testing was divided into two orthogonal parts:
1. a simple Simulink model that shows a scope of the input and the power spectrum of the signal.
2. a more complex GUI with data analysis.
The Simulink model communicates with the DAQ to acquire the data it is sampling from the analog circuit.
This data is:
1. shown in the scope as a differential channel
2. saved to a .mat log file
3. Analyzed by the spectrum scope and its power spectrum is shown in a graph.
The sample rate we used is 250 sample/sec in order to be able see clearly the EEG waves. We didnít use a higher sampling rate because it is not necessary for the waves we are interested in. The spectrum scope was configured to use a buffer (FFT length) of 4096 samples. To have a 3846 sample overlap between FFT calculations and to use a square window for the FFT.
In this experiment we had the subject tested for a total of 5 minutes. During the five minutes the subject was asked to close and open his eyes at specific times.
The results were analyzed using Matlab. As we shall see the actual results are in accordance with the expected results. One can clearly see the difference between the two graphs in the frequencies around 10Hz. When the eyes were closed, the power density of the alpha waves was about 5 times stronger than when the eyes were open, as expected.