We outputted a sine wave from the function generator at 9Hz and came up with the following plots using the following MATLAB code:
x = [ 
5333 
55354 
52449 
6093 
32767 
32767 
32767 
8453 
55195 
52430 
7259 
32767 
32767 
32767 
7356 
55003 
52422 
8294 
32767 
32767 
32767 
6573 
54890 
52453 
9227 
32767 
32767 
32767 
5892 
54785 
52461 
9899 
32767 
32767 
32767 
5232 
54684 
52454 
10570 
32767 
32767 
32488 
4680 
54588 
52451 
11259 
32767 
32767 
31989 
4065 
54512 
52494 
11959 
32767 
32767 
31536 
3527 
54430 
52537 
12559 
32767 
32767 
31071 
2980 
54343 
52575 
13303 
32767 
32767 
30446 
2162 
54228 
52639 
14426 
32767 
32767 
29509 
1061 
54059 
52807 
15771 
32767 
32767 
28598 
46 
53948 
53083 
17019 
32767 
32767 
27629 
64459 
53802 
53489 
18242 
32767 
32767 
26818 
63590 
53675 ]
t=linspace(0,.00004,100);
f=1./t
figure(1)
plot(t,x)
z=fft(x)
figure(2)
plot(f,z)
***The time interval is certainly incorrect for the values shown in figure 1 (amplitude vs. time)***
 *** We expect one large peak for the plot of the FFT of the array x. This is because an FFT transforms a function from time space into frequency space to show amplitude as a function of frequency.
The next goal in the process is to have the SDK code include lines that perform the FFTs directly in the hardware of the Zedboard, eliminating the use for a computer. The results of these can be used/combined for further analysis.


 
No comments:
Post a Comment