Need to find dominant frequencies, as well as peaks corresponding to noise and amplitude

Question

Nathan Jaqua 2019-9-30

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/482882-need-to-find-dominant-frequencies-as-well-as-peaks-corresponding-to-noise-and-amplitude

回答： Mahesh Taparia 2019-10-3

采纳的回答： Mahesh Taparia

doorbell.zip

Hi, I have 3 things I need to find -

What are the two prominent frequencies in the doorbell file? __________ and ______________

2. At what frequencies you see peaks corresponding to noise in spectrum? _____________ and ________________. Their amplitudes are ___________ and _____________

3.Plot the convolved spectrum. The new amplitude values of the noise are__________________________ and _________________.

I know I need to do 1 at y_spectrum, 2 at new_sound_array, and 3 at pwelch(Final_new_array). How do I do this with MATLAB?

clc;

close all;

clear all;

[y,Fs]=audioread('doorbell.wav');%reading audio file

y=y';

y_spectrum=pwelch(y);

sound(y_spectrum);%playing the spectrum using sound function

%% creating noise vectors

t=0:1/Fs:2;

t1=t(1,1:33292);

f1=4000;

a1=0.2;

x1=a1*sin(2*pi*f1*t1);

f2=5000;

a2=0.25;

x2=a2*sin(2*pi*f2*t1);

%% adding noise vectors to sound array

new_sound_array=y+x1+x2;

sound(new_sound_array,Fs);

new_sound_array_spectrum=pwelch(new_sound_array);

sound(new_sound_array_spectrum);

%%convolution of H array with new_sound_array

H=[0.0065 0.0103 0.0210 0.0382 0.0599 0.089 0.1032 0.1171 0.1221 0.1171 0.1032 0.0829 0.0599 0.0382 0.0210 0.0103 0.0065];

Final_new_array=conv(new_sound_array,H);

Final_new_array_spectrum=pwelch(Final_new_array);

sound(Final_new_array_spectrum);

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Mahesh Taparia 2019-10-3

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/482882-need-to-find-dominant-frequencies-as-well-as-peaks-corresponding-to-noise-and-amplitude#answer_394627

在 MATLAB Online 中打开

Hi,

The prominent frequency and amplitude can be estimated by plotting the power spectral density of the required signal. Since the signal which you have provided consists of some finite number of peaks, you can take the data tip cursor at the peaks in the respective plot and find the (x,y) coordinate. The abscissa will correspond to frequency of the signal and ordinate corresponds to amplitude. Adding the following code may help you:

figure;plot(y_spectrum) %plotting the spectrum 
figure;plot(new_sound_array_spectrum) %ploting the spectrum with noise 
figure;plot(Final_new_array_spectrum) %plotting convolved signal 