Announcement

Collapse
No announcement yet.

Designing Butterworth filter

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Designing Butterworth filter

    Hi all,
    I need to design a bi-directional sixth-order Butterworth filter with cut-off frequency of 6.0Hz to filter gait kinetic and kinematic data. I have found fdatool (filter design tool) in Matlab helpful. However, I do not know about some filter specifications such as Fs.
    Any comments regarding this would be appreciated.
    Regards
    Zahra

  • #2
    Re: Designing Butterworth filter

    Fs is usually sampling frequency. I'm sure about the matlab tool though. Butterworth filters are written and floating around the community, you might find someone who's already done this.

    Scott

    Comment


    • #3
      Re: Designing Butterworth filter

      You could also use the 'butter' function to create your filter, followed by the 'filtfilt' function to execute it.

      Wn=fc/(fs/2);
      [b,a] = butter(order,Wn,'low');
      output=filtfilt(b,a,input);

      Comment


      • #4
        Re: Designing Butterworth filter

        Hi Zahra,

        Fs is sampling frequency, the code Ryan has provided is a good example of how to implement it without FDTool. You just need to be aware that MATLAB uses a normalised cut off frequency where the cutoff frequency as a fraction of half the sampling frequency - hence the Wn = fc/(fs/2) portion of the code.

        If you are not comfortable with MATLAB you could also very easily write the filter in Excel. Check out David Winters book 'Biomechanics and Motor Control of Human Movement' for a detailed explanation of how to do this.
        Dan

        Comment


        • #5
          Re: Designing Butterworth filter

          Thank you Ryan and Dan,
          I tried that function and it worked well . But to make sure let me check fc=cut off frequency ( for my data is 6) and fs= sampling frequency (is 100hz) then Wn would be 0.12?
          Last edited by Zahra Safaeepour; June 7, 2012, 10:25 AM.

          Comment

          Working...
          X