Thread Tools
Oct 14, 2017, 02:22 PM
Registered User
mmormota's Avatar
The explanation about reflections and guard interval is not entirely good. There is possible overlapping of the direct and reflected signal regardless of the guard interval. This is an inherent feature as there are possible close and higher distance reflecting objects as well.
Of course the analog system is more sensitive, as a relative low amplitude reflection is visible too. However strong overlapping reflections cause misinterpreted symbols and data loss in digital transmission too.
I agree in the final conclusion that in most cases (using the system in open air far from reflecting objects like buildings etc.) circular polarization may not worth the bigger size, price etc. However in some cases (like ground vehicles, built-in area) it may worth the effort.
Sign up now
to remove ads between posts
Oct 14, 2017, 02:58 PM
Dethroned Knight of HD FPV
Thread OP
Quote:
Originally Posted by mmormota
The explanation about reflections and guard interval is not entirely good. There is possible overlapping of the direct and reflected signal regardless of the guard interval. This is an inherent feature as there are possible close and higher distance reflecting objects as well.
Can you describe how this reflected signal can create problems? Independently of everything else, with a 1/16 guard interval, this signal needs to travel 4 or 16 additional kilometres (depending on symbol length) before being received and interpreted. This assumes that somehow the signal manages to travel back and forth in a ping pong manner just before the receiving antenna, that is completely unrealistic. A much more realistic scenario is the signal bouncing off some *trees, houses or other objects nearby, get close to the receiving antenna, irradiate it for a while during the guard interval period and then leave for the horizon.

Quote:
Originally Posted by mmormota
However strong overlapping reflections cause misinterpreted symbols and data loss in digital transmission too.
Are you talking about indoor propagation with reflective objects and a strong transmitted signal? Because that's the only scenario that comes to mind where this might be a (theoretical) problem.
Last edited by fpv.blue; Oct 14, 2017 at 03:21 PM.
Oct 14, 2017, 04:00 PM
Registered User
mmormota's Avatar
Quote:
Originally Posted by fpv.blue
Can you describe how this reflected signal can create problems? Independently of everything else, with a 1/16 guard interval, this signal needs to travel 4 or 16 additional kilometres (depending on symbol length) before being received and interpreted. This assumes that somehow the signal manages to travel back and forth in a ping pong manner just before the receiving antenna, that is completely unrealistic.
It's kind of misunderstanding. Why do you think that guard interval saves against reflected signals?

You have a given packet length. Depending on the system, it could be from several microseconds to several milliseconds. The direct signal hits the receiver first, then (depending on the reflected signal path) the reflected signal hits it too. Typical delay is microseconds. This happens during the packet time. So the receiver receives the sum of the direct and reflected signal.

If the reflected signal is strong enough, the receiver can't correctly interpret the symbols of the direct signal, because of the additional strong interfering signal.

The circular polarised signal has the advantage, that a single reflection reverses the polarization (RHCP to LHCP and vica versa), it's the same as viewing a cork-screw in the mirror seems left handed. The receiver antenna has an attenuation factor to the reversed polarization, so the reflected signal amplitude becames lower, so the chance to interpret error-free symbols improving.
Oct 14, 2017, 04:29 PM
Dethroned Knight of HD FPV
Thread OP
Quote:
Originally Posted by mmormota
You have a given packet length. Depending on the system, it could be from several microseconds to several milliseconds. The direct signal hits the receiver first, then (depending on the reflected signal path) the reflected signal hits it too. Typical delay is microseconds. This happens during the packet time. So the receiver receives the sum of the direct and reflected signal.
Guard intervals aren't applied on every single packet, but on every single symbol instead. The receiver is "sleeping" during the guard interval, so reflected signals received during the guard interval are effectively having no effect on whatever the demodulator decided the received symbol was.
Oct 14, 2017, 04:45 PM
Registered User
mmormota's Avatar
Quote:
Guard intervals aren't applied on every single packet, but on every single symbol instead. The receiver is "sleeping" during the guard interval
Considering HD video transmission, you need several megabits per sec. So clearly impossible to choose so high guard intervals between symbols that is enough for avoiding the usual reflected signal delays.

1 M symbol rate means 1 usec between symbols, so guard time must be less then 1usec. In practice your symbol rate is higher than that.

Reflection delays are anything from zero to several microseconds. So the delayed version of the n. symbol easily hits the (n+1)...(n-k)th symbol.
Oct 14, 2017, 04:57 PM
Dethroned Knight of HD FPV
Thread OP
It might sounds like it's clearly impossible, but it's precisely what we are doing... I expect that as more reviews and more details for the product are released you will get a better idea of how that is possible. For now, yeah, just ditch those expensive and fragile antennas
Oct 14, 2017, 04:57 PM
g0t rabb1t?
ABLomas's Avatar
Quote:
Originally Posted by mmormota
Reflection delays are anything from zero to several microseconds. So the delayed version of the n. symbol easily hits the (n+1)...(n-k)th symbol.
Right.
Also, as already noted - reflected signal can directly interfere with direct signal, sometimes (when out-of-phase) corrupting signal/data, no guard can save information which isn't there (well, MIMO is different story).
But IMO for long range planes biggest issue is polarisation loss (reflections for long range, at least high altitude flights aren't big issue). Vee does not fix that completely. Airplanes with rudder/tail may do flat turns, but wings usually bank a lot in turn. You go with perfect signal, turn around to check what's happening and boom, suddenly you are flying blind (with analog video), just because your vertical antenna suddenly become almost-horizontal.

True-RC makes nice 1.2G antennas, incl. side-feed, which fits into wing nicely. Narrowband, but wide enough to be usable with analog and this system. With some tools and knowledge you can build even smaller antennas of the same type (with ultra-narrow frequency range and no, i won't tell you how) so size is not big issue now. And bye-bye polarisation loss, only nulls remains as issue (at least, for high altitude flights, here Vee may be better with dual-yagi on diversity).
Oct 14, 2017, 07:33 PM
Professional UAV Services
Redemptioner's Avatar
Quote:
Originally Posted by tta flyer
So multipath rejection isn’t important, what about cp’s other bonus, no polarization loss? Ie, banking the plane at long range causes more loss with linear antennas...is that still relevant?

Just curious since I already run a pepper box on the ground.
I use CP antennas mainly to prevent polarisation loss, I rarely fly where I would run into multipathing issues, compared to the price of this system (and staying away from VAS) a decent TrueRC antenna is cheap investment for buying some confidence in the antenna tune and more than likely improving the overall RF transmission and reception. It's like saying you bought some flash new sports car you plan to take on the race track, then cheap out on the tyres because race tyres cost more than 2nd hand tyres.....

Quote:
Originally Posted by fpv.blue
We are shipping a very nice vee antenna with this batch:

http://videoaerialsystems.com/portfo...w/vee-1280mhz/

The gain is 1.9 dBi, so it's not perfectly isotropic, but it should be close enough.
Well nothing from VAS is "very nice" I am sorry to say, the stuff coming from these guys is mass produced rubbish, even when they were "allegedly" making them by hand themselves they were terrible quality and looked like a "learn to solder class" had put them together and the "tune" isn't even close to a "quality" product. You are better off making your own dipole antenna rather than getting one from VAS, at least you can measure the lengths correctly and solder the connection on without half of china's solder stock pile every where....
Last edited by Redemptioner; Oct 14, 2017 at 07:39 PM.
Oct 14, 2017, 09:08 PM
Registered User
I wouldn't say VAS quality are crap. Been to 30km with my 1.2 pepperbox and mad mushroom from VAS. Only reason I turn back is because of battery capacity. Looking forward to some serious range test for this HD version
Oct 14, 2017, 09:48 PM
USA: LakeGeorge, New York
If you want QUALITY & PROVEN antennas strictly
for FPV........hire Hugo (TrueRC)!....or Tony (fpvLR)!

P.s.
Just dont hire Weinstein Company, he's a creep
Latest blog entry: My FAVORITE FPV CockPits'!!!!
Oct 14, 2017, 10:33 PM
Endurance, Altitude, and Range
daBKLYNdoorman's Avatar
Quote:
Originally Posted by rclab1
If you want QUALITY & PROVEN antennas strictly
for FPV........hire Hugo (TrueRC)!....or Tony (fpvLR)!

P.s.
Just dont hire Weinstein Company, he's a creep
+1 for TrueRC. Good antennas.
Oct 14, 2017, 11:35 PM
Registered User
sdigrazia1's Avatar
+1 TrueRc and Hugo!

Quote:
Originally Posted by daBKLYNdoorman
+1 for TrueRC. Good antennas.
Oct 15, 2017, 01:43 AM
Cruising at 399ft AGL
Bonafidepirate's Avatar
With all the talk about which antenna is best for what and why and who, don't forget that you're free to use whatever antenna you want... Standard SMA connectors mean the antennas you're using now with your analog system will more than likely drop right in place. That's exactly what I did, used the antennas that have proven to work best for me in my environment for the type if flying I do, and planned to subject this system to. I expect you all to do the same.
Oct 15, 2017, 02:40 AM
Dethroned Knight of HD FPV
Thread OP
The kit comes with receiving and transmitting antennas. We are not shipping directional receiving antennas, as those are too bulky and you can source them locally anyway. Get a patch if you want to go long range. We explained why above. People are saying what we are doing is impossible, well, it's actually happening. Get a patch. End of story.

If you have polarisation loss issues needing an antenna with gain lower than 1.9 dBi, you are probably flying fairly close, and even then it's probably a racing quad or similar. You don't want to mount a 1.2 GHz cp in there. Stick to linear, it won't break on the first crash.

I can confirm we were pleasantly surprised with the performances of the VAS vee antennas.
Oct 15, 2017, 03:39 AM
Registered User
Quote:
Originally Posted by fpv.blue
Unfortunately it looks like the UK will have to wait for a 2.4 GHz version of the system. Those regulations usually don't change overnight.
Would I be right in thinking that I couldn't use your 2.4Ghz version (when released) with a 2.4Ghz Spektrum DX9 control?

Shame you aren't thinking of releasing a 5.8Ghz system, your video looks better than connex (IMHO) and runs at 60fps with a WDR camera. Connex haven't offered that set up even with the latest HX camera (AFAIK). Really like WDR cameras


Quick Reply
Message:

Thread Tools