Assume I have only access to only one period of a signal that there is a transient effect on the signal as well.

I am pretty sure that there is a particular frequency in the signal. Is there a way to extract the exact value of DC and magnitude of that frequency by FFT (somehow erasing the effect of transient from the signal).

To make it more clear, check this figure:

enter image description here

I divided my signal into seven sections.

For each section, I want to know the DC value; and magnitude of the particular frequency that I always know exist in my signal.

My problem is that there is a transient effect on all sections that affect the next section for a short time. This phenomenon is much more clear in T6 in the figure.

Is there a way to get rid of the transient in signal processing?

  • $\begingroup$ Do you have some mathematical model of how you can describe the signal which you measure? To me currently it's not really clear, what you mean by "DC value and magnitude of the particular frequency". Do you mean your signal is something like $x(t)=a+b*sin(2\pi ft)+y(t)$ where $x(t)$ is the measured signal, $a$ is the DC, $b$ is the "frequency you know exists" and $y(t)$ contains the unknown transient function? $\endgroup$ – Maximilian Matthé Jan 11 '17 at 19:13
  • $\begingroup$ To summarize: do you basically want to do a DC-blocking filter to take care of the DC in T6 (avg level @ ~1.4)? Note that the whole signal has a DC offset. The transient appears to be is an effect of the discontinuity introduced by the T6 DC component (at the avg level of 1.4). $\endgroup$ – ruoho ruotsi Jan 11 '17 at 21:17

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.