I encounter a curious situation wherein the recording device of my EEG data appears to affect how many transients of definition "x" I find, especially strange since the amplitude criterion is relative (standard score instead of voltage).

One crucial difference is that the devices use different A/D resolution (12 vs. 22 bit) although they both encode the data in 16 bit (I already do not understand the difference between resolution and encoding, why both bit, but both different?).

To get to the point? How adjust for that difference? Also I excluded difference in sampling rate being crucial, because my current script adjusts for that based on two other devices I had to compare in the past, but which used the same A/D resolution...

Thank you dearly in advance!

  • $\begingroup$ 1 - What kind of transients? 2 - I suspect your 22-bit ADC is sigma-delta ADC. Sigma-delta ADCs will not react to a transient the same way as a SAR ADC would. 3 - TYpically, the data width should be equal or greater to your ADC resolution. So 12-bit ADC reading could be assigned to a 16-bit variable, without any loss of intormation. However, a 22-bit ADC would lose information if the reading is assigned to a 16-bit variable. $\endgroup$ – Ben Apr 15 at 12:12
  • $\begingroup$ Thank you! How would you advise to correct post-hoc for the wrong assignment? Indeed I am looking for transients in the sigma range. $\endgroup$ – Ivo Bulgermansky Apr 15 at 12:38
  • 2
    $\begingroup$ There's probably more going on here than you think. At this point, I think we should assume that you have two totally different data acquisition devices, one of which happens to be 22 bit and the other of which happens to be 12 bit. But we don't know what sort of anti-alias filtering is or is not there, or whether they have DC bias, or whether they have noise other than quantization noise, or how the signal is scaled to their range - etc., etc., etc. At the very least, can you collect two identical signals with each input device, and post the graphs here? $\endgroup$ – TimWescott Apr 15 at 15:22
  • $\begingroup$ That will not be easy :/ Due to the crisis I am remote from either device. $\endgroup$ – Ivo Bulgermansky Apr 15 at 15:30
  • $\begingroup$ Can you post plots from each device of similar types of signals, or post any plot that details what you are seeing? $\endgroup$ – Dan Boschen Apr 15 at 16:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.