1
$\begingroup$

I have sEMG signals, captured from commercially available device. the signals are stored in .tdms file format. i want to varify whether the captured signals are really sEMG (or simply EMG) signals or not? typical frequency range of EMG signals is 5 to 500 Hz and dominant frequency is 50 -150 Hz. can anyone guide me how may i find this from a recorded signal, using either MATLAB or LabVIEW.

$\endgroup$
  • $\begingroup$ By "dominant frequency" do you mean fundamental frequency? Or harmonic of the highest amplitude? $\endgroup$ – Florent Sep 6 '17 at 10:46
1
$\begingroup$

To detect the fundamental frequency of your signals, you can use self-correlation. The maximum of the self-correlation will correspond to your fundamental frequency (the second maximum, as self-correlation is always maximum at 0). You can then interpret the lag value of this secondary maximum to find the periodicity of your signal, and the amplitude of the maximum to evaluate the strength of periodicity. note : The value of the self-correlation at lag=0 is the energy of your signal. This can help as a scale for the periodicity strength.

Otherwise, to visually analyse the spectral content of your signal, you can just apply a Fourier Transform to it and plot the magnitude of the Fourier Transform.

$\endgroup$
  • $\begingroup$ To clarify the first statement: it is the time difference ("lag") of the first secondary maximum of the autocorrelation which indicates the fundamental frequency. The global maximum of the autocorrelation is always at lag zero. The value of this secondary maximum indicates the strength of the periodicity. $\endgroup$ – applesoup Jan 4 '18 at 13:31
  • 1
    $\begingroup$ Yep, thank you for the clarification, I'll edit my answer $\endgroup$ – Florent Jan 4 '18 at 14:13

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.