dump1090 resolves more a/c than VRS

Want to post something that doesn't quite fit into the other forums? This is the place for that.
Post Reply
T-EGJA1
Posts: 107
Joined: Sun Sep 08, 2013 12:11 am
Location: Mid-Devon, UK

dump1090 resolves more a/c than VRS

Post by T-EGJA1 » Fri Oct 10, 2014 11:30 am

As stated in the thread title, my browser display output direct from dump1090 (on a Ras Pi) shows more a/c on the map than VRS - see the screen-shots below (I've put red circles around the a/c that dump1090 shows that are 'missing' from the VRS display).

VRS display:
Image

dump1090 display:
Image

To clarify, the images above are crops from a single screen-grab from a dual monitor display; so are at the same instant in time.

My receiver 'Data Source' is set to 'AVR or Beast Raw Feed' and my 'Raw Feed Decoding' settings in VRS are as shown below:
Image

Have I screwed-up the VRS settings .... or is the discrepancy due to something else?
Terry

jonfear
Posts: 365
Joined: Sat Feb 09, 2013 12:15 pm
Location: Wick St Lawrence
Contact:

Re: dump1090 resolves more a/c than VRS

Post by jonfear » Fri Oct 10, 2014 1:29 pm

Hi Terry

You are not doing anything wrong. The way that various packages decode is as variable as the wind. VRS is quite conservative in the way it decodes data. It means that fewer planes are shown but they should be good data. VRS requires a few contacts before displaying an aircraft. My guess is that the dump1090 is rigorous about that.

I am sure that Andrew will make a few points when he next pop's onto the board.

Hope this helps

Jon
http://www.360radar.co.uk, the new name for MLAT Radar in the UK and Western Europe.

Former PP feeder Bm. No longer feeding. I do not have time to sort out imaginary problems with NTP
when it has been working fine for 2+ Years.

T-EGJA1
Posts: 107
Joined: Sun Sep 08, 2013 12:11 am
Location: Mid-Devon, UK

Re: dump1090 resolves more a/c than VRS

Post by T-EGJA1 » Fri Oct 10, 2014 3:10 pm

Thanks for your comments, Jon.

VRS being conservative in its decoding would certainly explain the discrepancy ... perhaps it may be just a little too cautious?

I've since compared the displays from both dump1090 and VRS, with adsbScope, which is running on a 'passthrough' rebroadcast from VRS - the number of a/c plots displayed by adsbScope is generally fewer than dump1090, but more than VRS, which fits with your observation about differences in ways of decoding the data.
Terry

milairuk
Posts: 34
Joined: Wed Apr 09, 2014 8:12 pm
Location: Chelmsford, Essex

Re: dump1090 resolves more a/c than VRS

Post by milairuk » Fri Oct 10, 2014 6:31 pm

Slightly off topic - I see on your VRS screen grab some markers - are these beacons / airports etc ??

How are these generated ?

John

T-EGJA1
Posts: 107
Joined: Sun Sep 08, 2013 12:11 am
Location: Mid-Devon, UK

Re: dump1090 resolves more a/c than VRS

Post by T-EGJA1 » Fri Oct 10, 2014 6:46 pm

Hi John,

The markers are airports - blue for civil and magenta for military. I generated them using the custom content plugin as described in this thread:http://forum.virtualradarserver.co.uk/v ... ?f=5&t=418 ... however it wasn't straightforward and caused me some considerable head-scratching; so let me know if you try it and run into problems, I may be able to help.

The symbols I've used were downloaded from: https://openclipart.org/collection/coll ... yNico/7797
Terry

milairuk
Posts: 34
Joined: Wed Apr 09, 2014 8:12 pm
Location: Chelmsford, Essex

Re: dump1090 resolves more a/c than VRS

Post by milairuk » Fri Oct 10, 2014 6:54 pm

Thanks Terry, I'll have a look and let you know if I get anywhere :geek:
John

agw
Posts: 2247
Joined: Fri Feb 17, 2012 3:20 am

Re: dump1090 resolves more a/c than VRS

Post by agw » Mon Oct 20, 2014 9:39 pm

By default VRS applies a slightly laxer version of the tests that are recommended in the ICAO specification. You can adjust the raw decoding settings to make it a bit more liberal and see how you get on with those.

The one likely to be useful without getting too many false positives is "Use local decoding for initial position fix" - if you tick / say yes to this then VRS doesn't wait for an odd and even position message within 10 seconds of each other before plotting the aircraft on the map, it'll take the first position message and decode it against your receiver's location. You need to specify a receiver location for this to work, if you don't it'll just ignore the setting.

The other thing to bear in mind is that VRS stops displaying aircraft that aren't sending messages sooner than some other programs. I can't remember the timeout off the top of my head but I think it's either 30 or 60 seconds. You can adjust that in the General options page.

Post Reply