How to plot Butterworth filters with own data?

Trying to use a bandpass butterworth filter on an ECG signal using the code below but whenever I use the lines:

h= filter( D,C,x) % where x= data

plot(h)

at the end of my code I just get a crazy plot which does not make sense or show my data. How could I accurately plot this filter with my data?

    fp= [3 120];
    fs= [2 200];
    fn= 1000;
    [N,fc]= buttord(fp/fn, fs/fn, 1, 60);
    [D,C]= butter(N, fc);
    %freqz(C,D)
    h= filter(D,C,x); 
    plot(h) % these last 2 lines give strange results

[Deepthi K]

How to get the values of filtered data. i.e. value of h?

Answer 1

[Star Strider]

It would help to know the sizes of ‘x’ and ‘h’.

You should also have a time vector, so:

figure
plot(t, h)
grid

will plot ‘h’ as a function of time vector ‘t’.

[Kathleen Nicole Rice]

x is a 26200x1 double (set of data from ECG) and hence h is the same size. There is no t vector due to a spotty data collecteor so t is just a repeating value from 1-2000 so I cannot use the t variable.

[Star Strider]

An improvement would be:

Fn= 1000;
Fs = Fn*2;
Wp= [3 120]/Fn;
Ws= [2 200]/Fn;
Rp =  1;
Rs = 50;
[n,Wp] = ellipord(Wp,Ws,Rp,Rs);                         % Elliptical Filter Parameters
[z,p,k] = ellip(n,Rp,Rs,Wp);                            % Elliptical Filter Design (Specify Lowpass)
[sos,g] = zp2sos(z,p,k);                                % Second-Order-Section For Stability

figure
freqz(sos, 2^14, Fs)                                  % Plot Filter Characteristics

h = filtfilt(sos, g, x)

figure
plot(h)

If you want to create a time vector:

t = linspace(0, 1, numel(h))/Fs;
figure
plot(t,h);

[Kathleen Nicole Rice]

Great, this fixed the problem. Thank you so much!

[Star Strider]

As always, my pleasure!