| IR Windowing and Schroeder curve |
 Posted: Sat Oct 24, 2015 8:10 am |
 |
|
| cekcim |
| Member |
|
| |
| Joined: 07 Sep 2015 |
| Posts: 11 |
| Location: Germany |
|
|
|
|
|
| Hi there,
I am doing RIR measurements with EASERA and I have some distortion components towards the end of my IR which I would like to window out. (I know the source of these distortions but it's practically impossible for me to remove them). The length of my IR is 11s and the distortion is between 10 and 11s. My question is when I put a window on this part of my IR, the Schroeder curve also changes its decay. Is this because of the backward integration of the IR? I find it quite strange because my IR decays to the noise floor after around 3s and the Schroeder curve equally represents this decay. When I apply the window, the curve bends to around 1sec.
Would really appreciate if someone could help out.
Thanks! |
|
|
|
|
| Posted: Sat Oct 24, 2015 8:40 am |
|
|
| thomas |
| Member |
|
| |
| Joined: 16 Mar 2011 |
| Posts: 73 |
|
|
|
|
|
|
| Hi,
You wrote, you put the window on the distorted part of the IR.
The part inside the window will be used for further calculations,
the part outside will be reduced (depending on the used window up to zero).
So You have to put the window on the part You want to use.
Start with a longer window, see the noise-tail in the schroeder-curve,
then reduce the length of the window stepwise (at the end of it) untill the noise tail dissapears.
Kind Regards
Thomas |
|
|
|
|
| Posted: Sat Oct 24, 2015 8:56 am |
|
|
| cekcim |
| Member |
|
| |
| Joined: 07 Sep 2015 |
| Posts: 11 |
| Location: Germany |
|
|
|
|
|
| Hi Thomas,
Thank you for your reply.
I tried what you said however exactly the same thing happened which I'm finding very strange.
Can it be due to the fact that I have Noise Compensation activated? |
|
|
|
|
| Posted: Sat Oct 24, 2015 5:16 pm |
|
|
| thomas |
| Member |
|
| |
| Joined: 16 Mar 2011 |
| Posts: 73 |
|
|
|
|
|
|
| Hi cekcim,
I think, it would be good to see Your IR, the ETC and the Schroeder-curve.
Could You send pictures of it? And perhaps a linkt to the IR for download (*.wav please).
Kind Regards |
|
|
|
|
| Posted: Mon Oct 26, 2015 10:12 am |
|
|
| cekcim |
| Member |
|
| |
| Joined: 07 Sep 2015 |
| Posts: 11 |
| Location: Germany |
|
|
|
|
|
|
|
|
|
| Posted: Mon Oct 26, 2015 2:52 pm |
|
|
| thomas |
| Member |
|
| |
| Joined: 16 Mar 2011 |
| Posts: 73 |
|
|
|
|
|
|
| Strange, my answer dissapeared! Second try:
Checked your IR - it's ok, unless the distortion at the end.
I suppose there is something wrong with your window settings.
Make sure:
- use a half window on the right side (I recommend a
rectangle window for RT calculations)
- use a window, that is long enough (a 3.5 s window worked fine
-> the noise correction wasn't necessary anymore; I get a T30 of
4.7 s (250 Hz oct.) and 1.6 s (8 kHz oct) as the longest and the
shortest value.
kind regards
thomas |
|
|
|
|
| Posted: Mon Oct 26, 2015 4:07 pm |
|
|
| cekcim |
| Member |
|
| |
| Joined: 07 Sep 2015 |
| Posts: 11 |
| Location: Germany |
|
|
|
|
|
| Thomas, thanks a lot for your help!
I applied a half-rectangle window with the right marker at 3.5s as you recommended and it seems like I got the same T30 results.
The problem with the weird looking Schroeder curve (as shown in the post-window link I sent you) seems to arise when I switch the Noise compensation on...not sure why..
Thanks again
Regards
Bernard |
|
|
|
|
| AFMG Network Forum Index -> AFMG EASERA |
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
All times are GMT
Page 1 of 1
|
|
|
|
|
AFMG Network Forum
