I applied Fourier transform on a sound (PCM generated digitally with sin function) and frequency detection is accurate. However, if I play the sound and record it, and then apply a Fourier transform to detect frequencies, my results are nowhere close to original results. The window size I have chosen for Fourier transform is 30 samples. ( Sampling frequency is 44100 Hz and 30 samples represent a symbol. Each symbol frequency is shifted by 1470 Hz.). Is my approach right ?
$\begingroup$
$\endgroup$
3
-
$\begingroup$ We need more information. How are you performing the fft (Matlab, Python)? Show us some code. What do the good and bad FFTs look like? $\endgroup$– David KMar 6 '14 at 16:51
-
$\begingroup$ How are you synchronizing your 30 sample frames with the recorded data? $\endgroup$– hotpaw2Mar 6 '14 at 18:24
-
$\begingroup$ I have fixed it. @hotpaw2 - Why do I need to synchronize the 30 sample frames ? Is it not sufficient if I check if the frequency exists (peak in its bin) ? $\endgroup$– user3388324Mar 7 '14 at 14:58
Add a comment
|
$\begingroup$
$\endgroup$
That's to be expected. If the input of your FF transform is 30 (real) points, you also get 30 real amplitudes out, for 30 different frequency bins. Each bin is (44100/2)/30 Hz wide, or about 700 Hz. That is quite wide. Add to that the effect of spectral leakage, which may quite well extend for 5 bins in either direction. 10 out of 30 bins will contain the ground frequency. Any higher harmonics will add to that.
Normally, people use much longer FFT's. 256 is quite a common choice, and 1024 isn't unheard of either.
