Transfer function of a system shows how reliably a system "transforms" its inputs into outputs. Imagine an electrode array that is recording signals from a saline solution with a sinusoidal signal generator in it and on each channel, there is a 5 degree phase advancement its signals due to faulty design. So the 20th channel would be 100 degrees ahead of the first channel and you may say: Oh look, I can see a physiological phenomenon in saline solution, I'm gonna publish it in nature!!
The same story is true when we are recording actual signals from brain tissue. Faulty electrode designs can mislead scientific research! so how to measure and control for this? that's what you'll learn in this tutorial.
Measuring transfer function of a headstage:
Just plug in the headstage tester and make it generate the white noise, record the data and keep it for analysis.
Measuring transfer function of a headstage + its electrode array:
There is a tiny BNC connector on the headstage tester, connect the BNC cable to it.
At the other end of the BNC, you have two alligator clips, connect the black one to your reference/ground on the electrode array.
Pour some saline into a baker and put the red alligator clip into the saline solution*.
put your electrode array into the saline solution and turn on the headstage, make it generate the white noise, record the signal and keep it for analysis.
Analysis:
Load the data into matlab and look at two signal characteristics:
FFT of the signal, ideally, all frequencies should have the same power.
Check the frequancy vs. phase plot, there should not be more than 5 degrees of phase shift between signals comming from different channels.
you may think that the distance between alligator clip and the electrode array may cause some sort of phase shift, but since the alligator clip is a huge signal source, it is unlikely that it happens.