1

I have a semi random noise signal, it does have some periodicity however.

I can estimate the power spectrum (of the blue signal, I thought that it would be more clear, because white noise is flat, but I digress)

enter image description here

How could I generate a noise signal from a power spectrum?

Brownian noise estimation is not the same as estimating from a power spectrum because of the 'sine wave' noise in the spectra, I would like to capture that.

Voltage Spike
  • 207
  • 2
  • 14
  • See https://dsp.stackexchange.com/q/322/11256 – MBaz Feb 29 '20 at 00:03
  • Yeah, I guess my problem is how do I get from a PSD (which I have) to FIR or IIR – Voltage Spike Feb 29 '20 at 05:33
  • I am trying to find a way to do this, but it would help to have some details. Making $1/f^2$ noise is no problem, but what exactly are the top two figures showing? The upper one seems to have a red trace under the blue trace. And what is the noisy sine wave? Is it supposed to be there or is it an an interference of some sort? Is the PSD for the third figure, with that little hump corresponding to the sine wave? – Ed V Feb 29 '20 at 14:54
  • @VoltageSpike Shouldn't you mention that in your question? Also, what is is Brownian noise related to your question? I'd like to ask you to edit your question to add more clarity. – MBaz Feb 29 '20 at 15:43
  • I would like to reproduce the blue signal – Voltage Spike Mar 01 '20 at 06:54

0 Answers0