Why is noise such a big problem for interpreting brain signals?

Basically, common mode noise signals have to be over 500 times greater than the brain signals to have a comparable amplitude on each channel. Further signal processing can reject that level of common mode noise by another factor of about 100 - by removing the common background signal between channels, for example by 'median filtering' - for each time sample, calculate the median value of all 14 channels and subtract that number form each channel value. In that way, any common signal detected across all channels is removed quite effectively.

The CMRR is around 85dB for electrical mains frequency interference and around 55dB across the working spectrum (Common Mode Rejection Ratio - how much crap common to both input terminals - reference and each EEG channel - makes it through the signal chain and appears as fake signal, 55dB is about 0.2%, 85dB is 0.005%).

Setting the EPOC to achieve GREEN sensors on all channels provides adequately low impedance for most purposes. Going over the top with wetting felts is a good plan for outside use since they will dry out quicker - the user can also add a few percent of household glycerin to the fluid which reduces evaporation rate and improves impedance.


Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request


Powered by Zendesk