Timeline for Calculating the fundamental frequency using Autocorrelation often gives half the expected value
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 20, 2016 at 14:32 | comment | added | PyroPez | Got it working properly now, thanks for all the help! I have one more question that I was wondering if you could answer: Do you know of any way to increase the accuracy of the fundamental frequency calculated by autocorrelation? It now consistently gives a result of 441Hz for a signal of 440Hz but ideally I'd like to get it as close as possible to the correct answer. | |
| Aug 19, 2016 at 23:22 | vote | accept | PyroPez | ||
| Aug 19, 2016 at 20:03 | comment | added | Marcus Müller | @PyroPez yes, it is! | |
| Aug 19, 2016 at 18:38 | comment | added | PyroPez | Sorry to ask so many questions (I'm not too knowledgeable when it comes to Signal Processing in case it wasn't obvious) but I can't seem to find anything on Google about time-inverse. Is it just reversing the signal? | |
| Aug 19, 2016 at 18:26 | comment | added | Marcus Müller | you can implement an autocorrelation that is faster than the time-domain implementation, but that is a multiplication in frequency domain, not a autocorrelation in frequency domain. | |
| Aug 19, 2016 at 18:26 | comment | added | Marcus Müller | you don't need to use the FFT to do autocorrelation; just convolve with the time-inverse; see wikipedia on autocorrelation. | |
| Aug 19, 2016 at 18:20 | comment | added | PyroPez | Thanks for the help. How exactly do you perform autocorrelation in the time domain? The code I'm currently using multiplies the square of the real and imaginary parts together but obviously that's not going to work if I don't take an FFT first. | |
| Aug 19, 2016 at 17:26 | history | answered | Marcus Müller | CC BY-SA 3.0 |