Filtre

[Sarhila123]

HI,

I have two similar signals, I want to filter the 'threshold' signal in such a way that the oscillations are not displayed, in order to compare the two signals afterwards. I used a low pass filter but it won't work. you can suggest me the right filter in my case. THANKS.

 

 

[GRCK5000]

Can you please attach your code?

[altenbach]

Are the two signals red and back, as displayed? Which one is the "threshold signal"?

Any lowpass filtering would round the sharp step.

What kind of "comparison" are you trying to do, and why the filtering?

Do you have a mathematical model, eg. Flat, then a sharp step followed by an exponential decay? Maybe you can just do a nonlinear fit on both and compare the parameters?

What is the experiment?

[Sarhila123]

as I have already indicated, there are two coline signals, I have the idea of ​​filtering one of the two then using it as a threshold signal for the other signal in order to count the small signal oscillations processed after when the counter equal 1 it is the value of systolic pressure and when the counter greater than or equal to a certain number of oscillations it is the diastolic pressure. but unfortunately I could not filter the signal in order to achieve its purpose.

[altenbach]

Ok, so you are interested primarily in the oscillations. Originally it sounded like you are wanted to remove them. So you only have one primary signal, but you want to remove the background and try to use filtering to create the background.

Can you attach a typical dataset?

If you are only interested in the data after the jump, maybe you could just fit it to a polynomial (using least absolute residual) or similar and subtract that.

[altenbach]

---

@altenbach wrote:

If you are only interested in the data after the jump, maybe you could just fit it to a polynomial (using least absolute residual) or similar and subtract that.

---

Here's how that could look like:

Of course there are many other ways to do all that.