When the Nyquist Frequency comes into play

[Loralynn M]

My question is: does the Nyquist Frequency (taking half the sampling frequency) come into play only after we have our raw recorded data? I.e. done on the computer afterwards. So, one would record at a certain frequency but then in order actually analyse the data we would need to take the Nyquist Frequency and use that as our data rather than the data at the original frequency we recorded it in.

[Simon]

Sampling happens when we convert from analogue to digital, so we need to worry about the Nyquist frequency then: when we record the raw data.

Our soundcard takes care of this for us. It includes a low-pass filter to remove all frequencies above the Nyquist frequency. This is done in the analogue domain, before going digital.

We also need to take care to do the equivalent thing when digitally downsampling any previously-recorded signal: our downsampling program must include a low-pass filter before the down-sampling step.

[Author]

Posts