Sharer Rankings - Use a different Ranking Approach?

akshay007

New Member
Hi all,

I noticed that the sharer rankings for ADS-B feeders (https://planefinder.net/coverage/receivers/all) are calculated solely based on the number of positions that each receiver reports. This means a receiver who has sent more positions will have it ranked higher than those who have sent less.

I see a potential flaw here: that this favours quantity over quality installations which have a wider range of coverage. This means one can live close to the airport and have poor coverage but due to the frequency of flights, will have them ranked higher, or, more importantly, this means one can easily rank higher if they share more positions (which can again differ due to the sources used) simply.

By contrast, all other major competitors use a combination of max range, avg. range and uptime to provide rankings.

For Example:
FlightRadar24 - as below

1650948271457.png


RadarBox: Uptime^2 * (Avg Range + (Max Range/5))

Due to this flaw, it may be possible to rank higher (as I have tested) by gaming the system by sharing more positions. This needs to change.

What are your thoughts on this, all?

Regards
Akshay
 
Hi Akshay,

Yes at the moment it is not ideal and most don't try and game the system so it does work. That said I can also see flaws with the other approaches also, in fact nothing is perfect.

For some time we have been working towards trying to make this fairer. One of the key pieces of this is using the data from the "Coverage Map" on the sharing site for individual users and assigning them a score that way and using that as a metric.

2022-04-26_08-03-42.png


This is based on the terrain modelling and in theory with a perfect setup is what a user should be able to achieve! This is harder to gain as we know what a user should in theory be able to achieve based on our terrain modelling.
 
Dear Lee,

Thank you very much for your reply! Indeed, by all means, each approach does have flaws and advantages of its own.
The coverage map approach is a good one rating the quality of an installation.

Good to know.

Thanks and regards
Akshay
 
My coverage area has been slowly getting smaller. Not sure why. Still see a lot of planes, since I'm close to the route from the EU into NYC, plus the Boston traffic. I used to see planes up in Canada and right down to NYC (during the winter, when the tree leaves aren't blocking).
1651855799440.png
 
From the Planefinder perspective, the value of the station to them must be a factor in any equation. After all, if a site is the only feeder within a 1000nm, you wouldn't want to put them off feeding.
 
I'll add to this old thread. Yes a change is still welcome. Look at the #1 ranked person. They have a relatively small range and undoubtedly very close to LAX. He is reporting a ton of ground traffic and receiving the position recognition. Planes on the ground moving slow from a gate to the maintenance hangar, overnight parking and back add no value to the system, but to the ranking stats, it's king.
 
Maybe if I put on a high gain LNA and a long 18 dBd gain Yagi pointed at Logan Airport in Boston? No, bad idea.
But, maybe a good 1090 band-pass filter and the LNA would be okay.
 
Maybe if I put on a high gain LNA and a long 18 dBd gain Yagi pointed at Logan Airport in Boston? No, bad idea.
But, maybe a good 1090 band-pass filter and the LNA would be okay.
What receiver are you using now? I'm having good performance with the ADSBx blue metal for 1090 and their orange for 978.
 
I was using a 1090 Puck (with Pi interface), but recently got a little black PlaneFinder box with a GPS antenna on a very long cable. So far, I'm finding it's better than the old one. That old Pi box was putting out too much RFI noise.
Now that the 100 ft trees in the backyard are starting to green up, my signals from Canada are going to disappear for the summer. Love GPS satellite display. So many GPS birds up there!
 
Back
Top