sampling and FFT of a sinusoid signal

Question

Yian Chen 2021-9-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1461189-sampling-and-fft-of-a-sinusoid-signal

评论： Chunru 2021-9-27

There is a 50HZ sinusoid signal g(t)=sin(2*pi*50*t) where 0≤t<0.1 and it is sampled at a rate of 250HZ, how can I plot this signal and 25 samples defined from 0-0.1s but does not inlude 0.1s? And how can I get the magnitude spectrum and phase spectrum of this signal?

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

Chunru 2021-9-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1461189-sampling-and-fft-of-a-sinusoid-signal#answer_795804

在 MATLAB Online 中打开

f0 = 50;

fs = 250;

t = (0:1/fs:(0.1-.5/fs)); % [0, 0.1)

g = sin(2*pi*f0*t);

plot(t, g);

L = 512

Y = fft(g, L); % compute spectrum, FFT length = 512

f = fs*(0:L-1)/L;

figure

subplot(211); plot(f, abs(Y)), title('Amplitude plot')

subplot(212); plot(f, (angle(Y))), title('Phase plot')

2 个评论
显示无隐藏无

Yian Chen 2021-9-27

编辑：Yian Chen 2021-9-27

Thanks for your answer, however, I don't exactly know why t = (0:1/fs:(0.1-.5/fs)) means [0, 0.1) and FFT length = 512, could you please explain them? Thanks!

Chunru 2021-9-27

t = (0:1/fs:(0.1-.5/fs)) means the time t starting from 0, with interval of T=1/fs, and eding at half-sample interval before 0.1 sec (to ensure that 01. is not included. Therefore this ensures t in in [0, 0.1).

In order to compute spectrum, you need to do FFT. By default fft(x) will take Fourier Transform with the number of frequency points equat to the data points (in your case it is 25). To have a smoother plot of the spectrum, we can use larger number of NFFT points, there fore we use FFT length = 512 above.

请先登录，再进行评论。