How do I get frequency response as values from multibandParametricEQ System object?

 채택된 답변

To extract frequency response, modifying or adding your own function to multibandParametricEQ.m might be helpful.
For example, you can use computefreqz, one of the dsp.util.FilterVisualizer object function to calculate frequency response within a multibandParametricEQ.
?
#1. Add following to visualize function defined in multibandParametricEQ.m
fVector = logspace(log10(20),log10(Fs/2),NFFT);
H = obj.visualObj.computefreqz(coeffs, fVector', Fs);
data = 20*log10(abs(H));
#2. Add output arguments to visualize function
function [data, fVector] = visualize(obj,NFFT)
#3. Confirm if modified method works
mPEQ = multibandParametricEQ(...
'NumEQBands',1,...
'Frequencies',9.5e3,...
'PeakGains',10);
[h, w] = visualize(mPEQ)
figure, semilogx(w,h)
ax = gca;
ax.YLim = [-25 25];
ax.XLim = [20 44100/2];
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
grid on
grid minor

댓글 수: 2

Thank you for your reply.
I tried to add these lines at the end of "function [data, fVector] = visualize(obj,NFFT)".
fVector = logspace(log10(20),log10(Fs/2),NFFT ); H = obj.visualObj.computefreqz(coeffs, fVector', Fs ); data = 20*log10(abs(H));
And then I got error like "not found computefreqz".
Something wrong? (I'm using R2016a. But other error occurred in R2018a. )
But I found I can get filter coefficients as below.
[B,A] = coeffs(mPEQ);
So now I can draw arbitrary graph such as ...
Fs = 48000; f = logspace(1,5,4096); f(f>Fs/2) = [];
h_all = 0; for i=1:size(B,2) sosPEQ = [B(:,i),[1;A(:,i)]]; h = freqz(B(:,i),[1;A(:,i)],f,Fs); h_all = h_all + 20*log10(abs(h)); end figure semilogx(f,h_all)
That's fine to me!
Thanks a lot.
That's great, thanks for sharing your experience with us.
&#01
The error you run into was because dsp.util.FilterVisualizer is not available on R2016a. This is just FYI, but you can use freqz instead like this;
fVector = logspace(log10(20),log10(Fs/2),NFFT);
H = freqz(coeffs{1}, [1;coeffs{2}], fVector', Fs);
data = 20*log10(abs(H));

댓글을 달려면 로그인하십시오.

추가 답변 (0개)

카테고리

도움말 센터File Exchange에서 統計と線形代数에 대해 자세히 알아보기

제품

릴리스

R2016a

질문:

2018년 8월 22일

편집:

2019년 10월 28일

Community Treasure Hunt

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

Start Hunting!