You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I can :
FFT.Windowing(real, BUF_SIZE, FFT_WIN_TYP_HAMMING, FFT_FORWARD);
FFT.Compute(real, imag, BUF_SIZE, FFT_FORWARD);
FFT.Compute(real, imag, BUF_SIZE, FFT_REVERSE);
FFT.Windowing(real, BUF_SIZE, FFT_WIN_TYP_HAMMING, FFT_REVERSE);
and get back the original data.
Without windowing I get back the original data.
If I try this with a HANN window some of the 1st and last values are undefined, or inaccurate.
Is this an issue with the inverse HANN windowing?
Or is it down to the nature of a HANN window?
See simple test below.. I have also tried with data ranging -128 to 128, 0 to 1 and -1 to +1.
I can :
FFT.Windowing(real, BUF_SIZE, FFT_WIN_TYP_HAMMING, FFT_FORWARD);
FFT.Compute(real, imag, BUF_SIZE, FFT_FORWARD);
FFT.Compute(real, imag, BUF_SIZE, FFT_REVERSE);
FFT.Windowing(real, BUF_SIZE, FFT_WIN_TYP_HAMMING, FFT_REVERSE);
and get back the original data.
Without windowing I get back the original data.
If I try this with a HANN window some of the 1st and last values are undefined, or inaccurate.
Is this an issue with the inverse HANN windowing?
Or is it down to the nature of a HANN window?
See simple test below.. I have also tried with data ranging -128 to 128, 0 to 1 and -1 to +1.
David
Hann
0 0.00 inf ak
1 1.00 1.023717 ak
2 2.00 2.048339 ak
3 3.00 3.021425 ak
4 4.00 4.000010
5 5.00 5.004950
6 6.00 6.008430
7 7.00 7.007216
8 8.00 8.004227
9 9.00 9.004525
...
246 246.00 246.007217
247 247.00 247.008926
248 248.00 248.008712
249 249.00 249.013732 ak
250 250.00 250.027526 ak
251 251.00 251.042877 ak
252 252.00 252.051025 ak
253 253.00 253.144409 ak
254 254.00 254.905548 ak
255 255.00 inf ak
Hamming
0 0.00 0.000048
1 1.00 1.000089
2 2.00 2.000482
3 3.00 3.000634
4 4.00 3.999955
5 5.00 5.000478
6 6.00 6.000649
7 7.00 7.000646
8 8.00 8.000188
...
246 246.00 246.000656
247 247.00 247.000656
248 248.00 248.000534
249 249.00 249.000732
250 250.00 250.000823
251 251.00 251.001190
252 252.00 252.000671
253 253.00 253.000747
254 254.00 254.000869
255 255.00 255.001068
The text was updated successfully, but these errors were encountered: