清空筛选条件
清空筛选条件

error matrix dimensions must agree designing yulewalk filter

1 次查看（过去 30 天）
Hamza Ashraf
Hamza Ashraf 2020-2-3
编辑： Hamza Ashraf 2020-2-4
Hi,
I am desiging a yulewalk filter to filter out the noise from input sound and detects the desired signal. Am attaching the code for function named designfilter that i am using in my code. It uses the magnitude values of the averaged noise frequency response in order to design the filter. But am having the error in code line
M=[-data(points)'-data(points2)'];
"matrix dimensions must agree"
can anyone help me out that where am doing it wrong
both the audio files as wav format are attached.
code for designfilter function
function [Hz,filter]=designfilter(finalnoise,freq,fs);
N=10;
const=fs/2;
%power spectral density
nfft=2^nextpow2(length(finalnoise));
pfs=abs(fft(finalnoise,nfft)).^2/length(finalnoise)/fs;
HpsdIS=dspdata.psd(pfs(1:length(pfs)/2),'Fs',fs);
%yulewalk filter
data=10*log10(HpsdIS.data);
points=round(linspace(1,230,230));
points2=round(linspace(231,512,50));
F=[(freq(points)*1000)/const (freq(points2)*1000)/const];
M=[-data(points)'-data(points2)'];
filter=struct('F',F,'M',M);
Hz=yulewalk(N,F,M);
end
here is my code where i am using designfilter function
[sound,fs]=audioread('noise.wav');
[wail,fs]=audioread('ambulancewail.wav');
bufferpos=0;
buffersize=5;
window=1024;
loop=round(length(sound)/window)-1;
%% noise buffer
noisebuffer(:,1)=sound((1:window),1);
for n=2:buffersize
noisebuffer(:,size(noisebuffer,2)+1)=sound((((n-1)*window)+1:n*window),1);
end
finalnoise=sum(noisebuffer,2)./buffersize;
%%
N = length(finalnoise);
xdft = fft(finalnoise);
xdft = xdft(1:N/2+1);
psdx = (1/(fs*N)) * abs(xdft).^2;
psdx(2:end-1) = 2*psdx(2:end-1);
freq = 0:fs/length(finalnoise):fs/2;
plot(freq,10*log10(psdx))
grid on
title('Periodogram Using FFT')
xlabel('Frequency (Hz)')
ylabel('Power/Frequency (dB/Hz)')
%% filter design
[Hz,filters(1).filter]=designfilter(finalnoise,freq,fs);

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Signal Generation and Preprocessing 的更多信息

标签

产品

Community Treasure Hunt

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

Start Hunting!

Translated by