Hi, I'm trying to perfom some analysis on a signal recorded from a load cell during an hammer strike. The strike is repeated with different materials hammer tips and at different velocities.
I would like to remove the noise, especially in the signals recorded for soft materials and small impact forces. I tried using an band pass filter (first figure) and a high pass filter (second figure) but both seem to amplify the noise in proximity of the peak.
I selected the frequences to cut looking at the FFT of the signal but changes if the thresholds don't make much difference
Are there any other ways of removing the noise from this kind of signal?

댓글 수: 5

Mathieu NOE
Mathieu NOE 2025년 2월 12일
hello
I don't see much noise on your pictures, but it would be nice to have some data if you can share them
what seems to me is that the sampling rate maybe a bit too low as we can see the signal having some straight portions in the "corners" (high curvature section)
now maybe another approach is to fit a model (gaussian pulse ?) to your data is you really need to smooth further.
Star Strider
Star Strider 2025년 2월 12일
If you are using the lowpass or bandpass functions, note that they design a FIR filter by default. If you want them to design a IIR filter, use the ImpulseResponse='iir' name-value pair. IIR filters usually produce better results, in my experience.
Mathieu NOE
Mathieu NOE 2025년 2월 13일
data ??
Riccardo
Riccardo 2025년 2월 13일
Hi, unfortunately i'm not allowed to share the dataset. @Star Strider your advice was very helpful thanks
Star Strider
Star Strider 2025년 2월 14일
My pleasure!

댓글을 달려면 로그인하십시오.

답변 (0개)

카테고리

제품

릴리스

R2024b

질문:

2025년 2월 12일

댓글:

2025년 2월 14일

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by