How can I plot fft ?

Hi,

I'm new student for writing in Matlab. so, I don't know how I can plot Y which is fourier transform of X(n). there is an error which says:"Vectors must be the same length".

x(n)=sin(100t)+sin(2000t)+sin(6000t);

my script is :

fs=6000;

% I consider 2 seconds for input signal

N=2;

t=(0:1/fs:N-1/fs);

w=(-N*fs/2:1/fs:(N-1)*fs/2);

X=sin(100*t)+sin(2000*t)+sin(6000*t);

X_fft=fft(X,fs);

X_fft=fftshift(X_fft);

Y=abs(X_fft);

figure;

subplot(3,1,1);

plot(t,X);

title('input signal X(t)before filtering')

%plot the first 100 samples

subplot(3,1,2);

plot(t(1:100),X(1:100));

title('Input signal with the first 100 samples')

%plot the DFT

subplot(3,1,3);

plot(w,Y);

title ('DFT for input signal befor filtering')

Answer 1

[Walter Roberson]

You have

 X_fft=fft(X,fs);

that tells fft to produce an output of length fs, 6000 . But your original data is length fs*N so you have problems when you try to plot(t,X) since t is length 12000 but X is length 6000

[Negar Sharafdoust]

I tried to solve the problem in the way you said and put N*fs instead of fs in fft(X,fs). but I got error. also I changed the w to get the result but it gives error in line plot (w,abs(X_fft)). can you tell me exactly which line I should correct?and how to write the command. thanks a lot

[Walter Roberson]

Your w is just completely the wrong size. Also, remember, it should be symmetrical, and remember that two adjacent frequencies are in each bin.

w = -N : 2/fs : N;
w(end) = [];

The second line there gets rid of an extra point as you would otherwise have an odd number (the same number for negative as positive, plus the entry for 0 exactly).