I have some center of pressure (COP) data obtained through Kistler force plate with 200 Hz frequency. Do I need to filter them as they are not such noisy? If yes why and what kind do you suggest more? Time series is like the figure attached.
I do not think the data are normal; based on what seems pink noise, i believe at least one force transducer had severed connection from piezo-sensor to amplifier.
Hello,
It depends on your objective, but mostly, investigation of the GRF is based on low frequency components and as I can see from the attached figure, the impact and active peak is clear, so I think a high pass filter (whit cut-off frequency between 40 to 50 Hz) would be enough.
Also, the active peak seems a little bit higher than the values reported in literature ( for example: Benno M. Nigg, The Role of Impact Forces and Foot Pronation: A New Paradigm)
what kind of shoe did you use?
subjects wore no shoes. what about the type of filter? Do you think highpass butterworth is fine?
The choice for window is depend on side lobes. In MATLAB toolbox you can see an illustration of the frequency response of the filter which you can choose the appropriate window based on the frequency response. Mostly, butterworth is enough.
Best Regards,
Hi Reza,
I have 3 references where they sample the COP at 100Hz but used 3 different filters. Prieto used a 4th order Butterworth low pass filter with a cutoff frequency of 5 Hz (Prieto et. al 1996 "Measures of Postural Steadiness: Differences Between Healthy Young and Elderly Adults") While Stins a 2nd order zero-lag Butterworth filter with 15 Hz or a 12.5 Hz of cutoff frequency (Stins et. al 2010 "To freeze or not to freeze? Affective and cognitive perturbations have markedly different effects on postural control", Stins et al 2009 "Sway regularity reflects attentional involvement in postural control: Effects of expertise, vision and cognition").
I think that you have to try with those frequencies and see your signal response.
Regards
Ivan
Clearly a lowpass, not highpass. The cut-off frequency should be between 5 and 8 Hz. I would start with a 5 Hz LP.
Sorry, I should correct my first answer. I meant a Low pass filter not High pass.
Thank you very much Ivan, Fernando and Arash.
I'm going to look at the refrences that Ivan has mentioned.
Thanks again
Thank you Marzieh
I have seen the syntax of the filter in Matlab. If I will have any question I will ask you.
- Badges
- Science topic