Timeline for All-pass phase compensation only in a specific frequency range
Current License: CC BY-SA 4.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 18, 2021 at 23:31 | comment | added | Dan Boschen | @skateskate Oh good thanks for the update and all makes sense. It was quite interesting as intuitively I wouldn't have seen by looking at the plot that the delay can be so far off from the linear slope (which would be the average delay essentially). Your waveform is effectively translating the signal in time very similar to how we can translate tones in frequency (by homodyning) but visually I don't see the complex rotation versus frequency that would do that but just may be a matter of scale on the plot. | |
| Feb 18, 2021 at 13:09 | comment | added | skateskate | Thank you again @Dan Boschen, yes it seems that increasing the time span of the filter solves the issue. I am pretty busy right now and I don't think I have time to post an update with the new results, I hope I can add them in the following days. | |
| Feb 17, 2021 at 14:25 | comment | added | Dan Boschen | @skateskate please let us know if increasing the time span resolved this. | |
| Feb 17, 2021 at 8:19 | vote | accept | skateskate | ||
| Feb 17, 2021 at 2:29 | history | edited | Dan Boschen | CC BY-SA 4.0 |
added 190 characters in body
|
| Feb 16, 2021 at 20:51 | history | edited | Dan Boschen | CC BY-SA 4.0 |
added 185 characters in body
|
| Feb 16, 2021 at 18:17 | comment | added | Dan Boschen | @skateskate Very interesting problem. I played with it further and have updated conclusions which I added to my answer. Not sure how easily you can increase the time span of your equalizer but that could also be potentially done at a lower sampling rate to keep the processing level the same (depending on your overall waveform bandwidth such that you don't then end up with frequency domain aliasing). | |
| Feb 16, 2021 at 18:16 | history | edited | Dan Boschen | CC BY-SA 4.0 |
added 3274 characters in body
|
| Feb 16, 2021 at 13:21 | comment | added | skateskate | Ok, I have edited (EDIT 2) my question. | |
| Feb 16, 2021 at 12:58 | comment | added | Dan Boschen | Great- in the meantime can you update your “edit” section to include the impulse response right after your initial waveform where you show the linear fit using the exact same methodology as you do further below with the modified waveform? | |
| Feb 16, 2021 at 12:50 | comment | added | skateskate | Yes, here is the phase vector "ph" from the code: drive.google.com/file/d/1JyP3ngNMxlfvA_Gqjy1I5BzeSRFptbOI/… | |
| Feb 16, 2021 at 12:00 | comment | added | Dan Boschen | Interesting; could you plot your impulse response following the same computation prior to the mask? Do you have a way to provide a link to your actual data? | |
| Feb 16, 2021 at 11:57 | comment | added | skateskate | Thank you very much, I have edited my question with the Octave code I am using, implemented following your suggestions (with which unfortunately I have aliasing). | |
| Feb 16, 2021 at 3:24 | history | edited | Dan Boschen | CC BY-SA 4.0 |
added 56 characters in body
|
| Feb 16, 2021 at 3:18 | history | answered | Dan Boschen | CC BY-SA 4.0 |