2nd Order Digital Butterworth filter

Hi Guys,

 I need help designing a 2nd order low pass Butterworth filter with a frequency cut off of 12 Hz that is to be bi-directionally filtering the data.

How do I go about designing a filter like this in Matlab?

   I realize there is a function 'butter' to design a lowpass filter, but was kind of confused on applying it to my matrix that has about 9 columns and 30,000 rows, and I need to apply the filter for every row.

   Thanks for your help in advance!

[Jan]

We need the frequency of the signal also.

[Daniel Shub]

You want to filter signals which are 9 samples long?

[Jan]

I'm sure "9 columns and 30'000 rows" mean "columns of length 9 and rows of length 30'000".

[Aman]

yes I meant 9X30000 matrix

Answer 1

[Jan]

[B, A] = butter(2, 12 / (Freq * 0.5), 'high');
filtered = transpose(filtfilt(B, A, transpose(Signal)));

Freq is the frequency of your signal. If you need a low-pass filter change the 'high'. See help butter on demand.

[Aman]

what exactly are B, A? From the matlab help it shows B & A as the filtered n+1 row vectors but I don't really understand what that implies? Say I have a matrix N that is 9x6400, how would I implement this into the butter code?

[Honglei Chen]

Designing the filter is independent to your signal. Once the filter is designed, you can filter your signal using filter function, such as filter(B,A,Signal)

[Aman]

ah ok that makes much more sense, thank you