0
$\begingroup$

I have one recorded signal (.wav) and the calibration signal (.wav - 94dB @ 1kHz).

I want to know how to calibrate the recorded signal using the calibration signal, for example, in Matlab. My tasks are

  • estimating background noise level,
  • equivalent levels
    • Aeq,
    • Ceq,
    • Zeq,
  • max-values of
    • S and
    • F time
  • integrations
    • ASmax,
    • AFmax and
  • peak values
    • Apeak,
    • Cpeak,
    • Zpeak.
| improve this question | |
$\endgroup$
  • $\begingroup$ so what do you want to come out of your MATLAB program? an envelope in dB? $\endgroup$ – robert bristow-johnson Dec 1 '16 at 18:32
  • $\begingroup$ my tasks are estimating background noise level, equivalent levels (Aeq, Ceq, Zeq), max-values of S and F time integrations (ASmax, AFmax) and peak values (Apeak, Cpeak, Zpeak). $\endgroup$ – rfbrgs Dec 1 '16 at 18:35
  • $\begingroup$ Add that info to your question instead of as a comment!! $\endgroup$ – Marcus Müller Dec 1 '16 at 18:55
1
$\begingroup$

First you need to know something about your calibration signal. Lets say you know that it is a 1kHz sine with an RMS amplitude of 94dB SPL. If you are confident that this is all your calibration recording includes (which is likely), then your calibration factor, is the difference between the actual dB SPL level and the 'digital' dBFS level. In matlab code, it looks like this:

caldb = 94 - 20*log10(rms(calibration_signal);

It could be that the calibration level is specified as peak SPL, then you would just substitute the peak value for the rms value above.

Now you can calculate all your values (LAeq etc) in dBFS (i.e. -inf to 0dB) and simply add the calibration factor, to get a calibrated SPL.

| improve this answer | |
$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.