Author Topic: OGN-R uplink typical range  (Read 12639 times)

buzz53

Re: OGN-R uplink typical range
« Reply #15 on: August 19, 2020, 12:50:10 pm »
I'm doing various range tests, ground and air, in both directions to try to figure out why I see relatively poor performance despite what I think should be a good antenna installation. I thought I should first check my bridge is working OK. According to my test gear, mine is outputing a little over 200mW. Is this correct? I imagine it is as we should be allowing for some antenna gain to remain within the 500mW ERP limit.

TIA
Alan

buzz53

Re: OGN-R uplink typical range
« Reply #16 on: August 22, 2020, 09:28:46 am »
OK, so the power output is a big secret! Another question:

A few people have kindly sent me track files in order to evaluate their uplink performance. One set in particular is surprising as the installation is known to have a good downlink but the uplink is apparently dire. Pending a test flight, could someone tell me whether the appearance (or not) of a ground station in the track log, and hence on the Aircrew playback, is affected by the horizontal and/or vertical filter settings for aircraft targets? Similarly, the appearance or not of tower symbols and the OGN-R label on the radar display?

If they are filtered from the log, I would suggest that they should not be, as this precludes post flight analysis.

Another suggestion: I just realised that the log doesn't seem to include the ownship ID, is that on purpose? It seems it would be handy to have.

Alan

JCurtis

Re: OGN-R uplink typical range
« Reply #17 on: August 22, 2020, 10:26:19 am »
I have recently updated some of my lab equipment, so can cover RF up to 6GHz now.  This enables real time and swept frequency analysis, plus all the normal power measurements etc.  Quite interesting now being able to "watch" frequency hopping systems working in real time.

I guess I could renew my PAW licence and test a unit I have here, unless Lee would like to issue me a short term testing licence?
Designer and maker of charge4.harkwood.co.uk, smart universal USB chargers designed for aviation.  USB Type-A and USB-C power without the RF interference. Approved for EASA installs under CS-STAN too.

Admin

Re: OGN-R uplink typical range
« Reply #18 on: August 22, 2020, 10:27:21 am »
Hi Alan,

Unfortunately the data in the track file is not nearly sufficient for the analysis you are trying to do, internally we have been adding detailed telemetry data for communications between ground/air, and behind the scenes the dev team is working vigorously to categorise the disparate set of installations and rationalise why some installations seem to undergo RF asymmetry, we have some ideas, but much too early to make any kind of definitive statement.

The real benefit here is that data is King, and we have harnessed more data than you could shake a stick at in the months since the launch of the ATOM/GRID network. In fact not only for PilotAware equipped aircraft, but additionally for all those Transponder, Flarm and cap1391 devices.

Nonetheless, we are actively working to constantly improve performance of the system, and there are many ways to do this by changes to software, hardware and even at the system level.

The work that is ongoing is both challenging, and exciting - and as my good friend Alan Walker told me many years ago "You will solve these problems ..."

Thx
Lee

buzz53

Re: OGN-R uplink typical range
« Reply #19 on: August 22, 2020, 10:33:34 am »
Lee,

Thank you, but I would still be grateful for a answer to the question! Am I being unreasonable? Like the power question it can be determined by experiment if necessary.

Alan

Admin

Re: OGN-R uplink typical range
« Reply #20 on: August 22, 2020, 11:58:55 am »
Hi Alan
The design of PilotAware meets ETSI EN300-220 at 27dbm.

If you think there is an issue, we will happily investigate, if there is a fault, and product is still under warranty this will be free of charge
If out of warranty, then there will of course be a charge for labor, and any necessary parts

you can send an email to support@pilotaware.com, and we can provide further information on costs and return information

Support Team

JCurtis

Re: OGN-R uplink typical range
« Reply #21 on: August 22, 2020, 02:38:03 pm »
I'm doing various range tests, ground and air, in both directions to try to figure out why I see relatively poor performance despite what I think should be a good antenna installation. I thought I should first check my bridge is working OK. According to my test gear, mine is outputing a little over 200mW. Is this correct? I imagine it is as we should be allowing for some antenna gain to remain within the 500mW ERP limit.

TIA
Alan
With the caveat that all of this depends on the accuracy and calibration state of your test equipment...

200mW is 23dBm.  So a little over, say 225mW, is 23.5dBm.  Add the supplied antenna gain, which I think is 3dBm, we get 26.5dBm.  The EIRP limit is 27dBm so that looks about right.  Naturally there are also cable losses too, the measurement method, and the accuracy and calibration of the equipment all come into play.  I'm assuming you have measured any loss of the DC block, pad, the cables used etc. at at that frequency too and taken them into account?

On the basis of 'a little more than 200mW' which is assumed is direct from the output of the PAW itself, that PAW would seem to be fine.  I also presume the frequency was OK too, I guess it might have drifted a little over time?
 
Perhaps measure the RF output at the antenna connection point on the aircraft itself?
Designer and maker of charge4.harkwood.co.uk, smart universal USB chargers designed for aviation.  USB Type-A and USB-C power without the RF interference. Approved for EASA installs under CS-STAN too.

buzz53

Re: OGN-R uplink typical range
« Reply #22 on: August 23, 2020, 10:02:54 am »
JC: I agree with your numbers and conclusion, thank you. My trusty HP8594E has not been calibrated for a very long time, but it gives the "expected" answers for my FLARM, SoftRF and various balloon telemetry units I've built myself so I am pretty confident in it. Be good to get confirmation though, if you get a chance.

Lee: the question I was referring to was about the filtering of groundstations by range/height difference in the log. My filters are set very broad at the moment and I appear to see a log entry for every packet received from groundstations which is surely all that is required to fully analyse the performance of my own equipment? What am I missing? However if the groundstations are filtered, as I suspect,  then that would explain why I see such poor uplink results on other peoples' logs and why I suggested they should not be filtered.

Alan

JCurtis

Re: OGN-R uplink typical range
« Reply #23 on: August 24, 2020, 01:05:05 pm »
JC: I agree with your numbers and conclusion, thank you. My trusty HP8594E has not been calibrated for a very long time, but it gives the "expected" answers for my FLARM, SoftRF and various balloon telemetry units I've built myself so I am pretty confident in it. Be good to get confirmation though, if you get a chance.

Alan

Thanks to Lee for getting me up and running again.

I get 22.45 dBm at 869.526733MHz from my *very* old bridge board.  But I have -1.03dB of loss in my cable, DC block, and 20dB PAD at that frequency.  So the adjusted power is 23.48 dBm.  Pretty much as expected really.

If I assume it's a 25kHz wide channel (max allowed under the ETSI spec), the total power available in the channel is 27dBm for each ~15ms transmission pulse every ~1.7s - so looks to be exactly to spec.  The 3dB "gain" of the end fed dipole will direct more RF energy to a more horizontal plane.  After all an antenna doesn't amplify a signal, the gain is all to do with directing more of the fed RF power compared to an Isotropic antenna.  For the end fed di-pole, this means the RF pattern is more of a ring donut shape than a sphere.

I'm not a RF bod by any means, that is a very specialised area.  I can just about keep up with what I need to for the projects I'm involved in.
Designer and maker of charge4.harkwood.co.uk, smart universal USB chargers designed for aviation.  USB Type-A and USB-C power without the RF interference. Approved for EASA installs under CS-STAN too.

buzz53

Re: OGN-R uplink typical range
« Reply #24 on: August 24, 2020, 09:52:56 pm »
JC: Super, thanks, so all as expected on the power front.

On the filtering issue, I had a test flight today (actually 2 flights, as when I tried to change the config in the air the dreaded "Http: Get request method not supported" struck again. I had a lot of trouble with that initially but it seemed to have gone away).

I'm now pretty sure that:

1) Range filter settings do NOT affect the label showing the number of OGN-R stations displayed.
2) They DO control whether ground stations are displayed on the RADAR screen.
3) They DO control whether the ground station uplink status is recorded in the track file and hence are available for post-flight analysis.

I think (1) is obviously fine, (2) is a matter of taste, but (3) is surely not what's wanted?

Despite not changing anything, I do now seem to be getting better range performance, although it is inconsistent. I'll write more later when I have gathered more data.

Alan

rogerabc

Re: OGN-R uplink typical range
« Reply #25 on: August 25, 2020, 10:19:05 am »
Thanks Alan, that's interesting news & I hope this behaviour can  be changed in a future version.

Whilst accepting there are other points of view, I now post mine as feedback to the PAW development team:

Until EC is mandated I prioritise look out & listen out, so audio warnings are my primary PAW channel (with a mute switch), I do not want to monitor the radar screen which detracts from lookout.

Continuous ATOM grid reception will maximise my EC detection capability and I was trying to monitor this using aircrew.co.uk track analysis which lead me to believe that my uplink capability was dreadful (I had my filters set to close range detection).

Please would the team consider other ways for users to monitor their uplink status - ideally for me the option for a discrete audio prompt, perhaps every 5 minutes,  "Atom".

Perhaps there's a better way & perhaps I'm a minority but it's constructive feedback which is a good thing!

The atom grid really is the key to effective cross platform EC.

Thanks to the PAW team & best regards to all,

Roger


Admin

Re: OGN-R uplink typical range
« Reply #26 on: August 25, 2020, 11:42:26 am »
JC: Super, thanks, so all as expected on the power front.

On the filtering issue, I had a test flight today (actually 2 flights, as when I tried to change the config in the air the dreaded "Http: Get request method not supported" struck again. I had a lot of trouble with that initially but it seemed to have gone away).

Urgh, this is Android right ?
An Android update came out which caused this - it is fixed in the release to be put out literally within a week or two.

Quote
I'm now pretty sure that:

1) Range filter settings do NOT affect the label showing the number of OGN-R stations displayed.
2) They DO control whether ground stations are displayed on the RADAR screen.
3) They DO control whether the ground station uplink status is recorded in the track file and hence are available for post-flight analysis.

I think (1) is obviously fine, (2) is a matter of taste, but (3) is surely not what's wanted?


Regarding the track file, it is not really intended as a full diagnostic tool.
We are adding lots of telemetry data in the next release to help do that across the board.

Quote
Despite not changing anything, I do now seem to be getting better range performance, although it is inconsistent. I'll write more later when I have gathered more data.
Obscuration plays the biggest part here which is what we have determined.
If you PM me with your flight details I can take a look at our records

Thx
Lee

buzz53

Re: OGN-R uplink typical range
« Reply #27 on: August 25, 2020, 12:33:26 pm »
Quote
Urgh, this is Android right ?
An Android update came out which caused this - it is fixed in the release to be put out literally within a week or two.

Yes, so that's great news, although it's a 2013 Nexus 7 with a very old Android version so maybe not the same issue. Look forward to the new release.

Quote
Regarding the track file, it is not really intended as a full diagnostic tool.
We are adding lots of telemetry data in the next release to help do that across the board.

It may not be the intention, but if you did then it would be a very powerful diagnostic tool and I can't see any downside at all. How else is one to assess what's going on other than by staring at the screen during the flight? Will your new telemetry be available in detail to each user for their own aircraft or will it be for internal use for overall statistical purposes? If not then that would of course still be very interesting from a system point of view, but really serves a different purpose and is presumably dependant on the downlink. It seems quite a fraught way to achieve something that would be directly available in the log file for each user.

Quote
Obscuration plays the biggest part here which is what we have determined.
If you PM me with your flight details I can take a look at our records

Thanks, will do.

Alan

Admin

Re: OGN-R uplink typical range
« Reply #28 on: August 25, 2020, 02:55:18 pm »
Quote
Regarding the track file, it is not really intended as a full diagnostic tool.
We are adding lots of telemetry data in the next release to help do that across the board.

It may not be the intention, but if you did then it would be a very powerful diagnostic tool and I can't see any downside at all.
Support mostly, our resources are scarce, and we must be very careful where we spend them

Quote
How else is one to assess what's going on other than by staring at the screen during the flight?
Automation of data collection and processing, followed by data representation.
it is not viable to have anybody trawl through megabytes of text logfiles trying to understand the meaning.

Quote
Will your new telemetry be available in detail to each user for their own aircraft or will it be for internal use for overall statistical purposes?
initially internally, but the goal is to use the data in order to provide feedback to users of installations where we can clearly see something is not representative. This includes all sorts of things such as the S/W version being run to the quality of the bidirectional datalink

Quote
If not then that would of course still be very interesting from a system point of view, but really serves a different purpose and is presumably dependant on the downlink. It seems quite a fraught way to achieve something that would be directly available in the log file for each user.
For the beta released currently being trialed it is working very well across the entire network, we are gathering very valuable information.

There are a few dedicated individuals (such as yourself) who are prepared to invest time and effort into the detailed technicalities of what is going on by trying to analyse the data from the perspective of the individual installation, that is the exception, and not the rule. We need to be able to carry out these tasks using a system level approach, requiring no intervention from the end user.
This is exactly what is happening now, for those individuals flying with the beta release, and will be extended when this goes live.

thx
Lee




exfirepro

Re: OGN-R uplink typical range
« Reply #29 on: August 25, 2020, 03:10:27 pm »

Until EC is mandated I prioritise look out & listen out, so audio warnings are my primary PAW channel (with a mute switch), I do not want to monitor the radar screen which detracts from lookout.

Continuous ATOM grid reception will maximise my EC detection capability and I was trying to monitor this using aircrew.co.uk track analysis which lead me to believe that my uplink capability was dreadful (I had my filters set to close range detection).

Please would the team consider other ways for users to monitor their uplink status - ideally for me the option for a discrete audio prompt, perhaps every 5 minutes,  "Atom".

Perhaps there's a better way & perhaps I'm a minority but it's constructive feedback which is a good thing!

The atom grid really is the key to effective cross platform EC.

Thanks to the PAW team & best regards to all,

Roger

Hi Roger,

If you are talking about the 'Positional Contacts Settings' Range Filters in PilotAware, these will certainly affect the display of Ground Stations with the current versions of PAW and SkyDemon software, because SD has historically treated PAW Ground Stations as 'Traffic', but developments are in hand which will significantly improve this situation. In any event, these particular filters are only intended to be used with display software which doesn't incorporate its own horizontal or vertical traffic filters. For systems such as SD, which do, these filters should be left at the default - 'Display All' - and the filters in the display software used to determine what is (or isn't) displayed.

Hope this helps.

Best Regards

Peter