Timeline for How were windows originally conceived?
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 31, 2019 at 12:45 | history | edited | Matt L. | CC BY-SA 4.0 |
added 42 characters in body
|
| May 21, 2013 at 23:46 | comment | added | hotpaw2 | If your signal is properly bandlimited, then the only way more sensors would help is if they captured a longer signal. A longer signal would allow the use of a window of greater length, which would allow a narrower main lobe. | |
| May 21, 2013 at 18:49 | comment | added | TheGrapeBeyond | @MattL. Ok, I will think about it some more and make new q. Thank you. | |
| May 21, 2013 at 17:50 | comment | added | Matt L. | Not sure if I understand what you mean. You might want to work this out a bit and post a new question here. | |
| May 21, 2013 at 16:07 | comment | added | TheGrapeBeyond | @MattL. I see. Finally, do you suppose something like what I am saying might be possible if additional sensors were brought to bear on a given signal? Thanks once more. | |
| May 21, 2013 at 15:58 | comment | added | Matt L. | Do you know filter design? It's the same thing, you can have a higher stopband attenuation if you allow a wider transition band; if you need a narrow transition band then your stopband attenuation will be lower, given a fixed filter order. There's even a trade-off between maximum sidelobe level and sidelobe energy; minimizing one of them will give a relatively large value for the other one. | |
| May 21, 2013 at 15:54 | comment | added | TheGrapeBeyond | @JasonR Might you suppose both can be optimized simultaneously with additional information, (such as from additional sensors)? | |
| May 21, 2013 at 15:54 | comment | added | TheGrapeBeyond | @MattL. But why not? I guess my question is why are those two contradictory criteria. | |
| May 21, 2013 at 15:34 | comment | added | Matt L. | No, there's a trade-off and you can choose the point of the trade-off curve that you want to be on. You basically cannot get everything at the same time. | |
| May 21, 2013 at 15:34 | comment | added | Jason R | @TheGrapeBeyond: When you have two attributes that trade off with each other (i.e. making one better makes the other worse), then you can't optimize them both at the same time. As the designer, you have to pick an implementable point in the trade space. | |
| May 21, 2013 at 15:18 | comment | added | TheGrapeBeyond | Even just skimming it I can tell this is already a fantastic paper, thank you for that. I suppose my 'ad-hocy' suspicion of how most windows were designed is true then. (Visual smoothness in the time domain). Question: Is there a kind of 'limit' as to the most ideal you can make a window, such that its mainlobe width is as small as possible, while its sidelobe levels are as attenuated as possible? | |
| May 21, 2013 at 14:57 | history | answered | Matt L. | CC BY-SA 3.0 |