I am using matlab wavelet toolbox to denoise physiological signals, I am plotting the denoised signal on top of the original noisy signal and making sure spikes were not removed as a measure of effectiveness as well as plotting the FFT of the two signals (original and denoised).

Is this a correct way to check my results of DWT denoising? I saw some paper using mean squared error and SNR as a check.

  • $\begingroup$ "correct": well, that depends on what you want to use that effectiveness measure for! $\endgroup$ Jul 13 '20 at 21:24
  • $\begingroup$ I want to make sure my signals of interest are not affected by the denoising rather only the noise is flattened or attenuated $\endgroup$
    – Shannon
    Jul 14 '20 at 5:23
  • $\begingroup$ Well, I guess if your signal of interest is these spikes? $\endgroup$ Jul 14 '20 at 8:47
  • $\begingroup$ I am still working on appropriate quality metrics for sparse signals, after all these years $\endgroup$ Jul 15 '20 at 22:18

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.