I’m considering building my own FlightAware PiAware ADS-B Ground Station.
I live in Nome, Alaska and internet bandwidth is very limited. For example, I pay $165 for 100gigs of data a month. If I’m running a basic Pi setup like the one shown in the FlightAware videos how much bandwidth will I consume each month to operate the system?
Definitely, in the big scheme of things we don’t have much traffic. But there is a fair amount of local traffic going back and fourth between the hub here in Nome and the surrounding villages. Not to mention the bigger jets en route to/from Asia that fly overhead daily.
I can probably do it knowing that it will take less then 5 gigs a month. But my wife and kids would absolutely kill me if I used up 10/20/30 gigs a month for my hobby.
In Norfolk, VA area, close to an international airport, and under a busy air corridor, I had 2-3GB/month with MLAT on. Disabling that feature reduces the usage with 30-40%, for me was around 2GB.
But it all depends of how “far” your antenna can see. Mine didn’t go past 40 miles out.
If you plan to feed other trackers (like FR24), the usage doubles again.
For Mlat, you need other feeders in your area, so maybe it does not work anyway?
Then, here in Europe a lot of planes have ADS-B, my impression is that it is different in the US, and it depends on the type of aircraft. This could mean without MLAT you might miss more than just a little traffic?
You could check the site of the person already feeding in your area, to have a rought idea about traffic.
If you view anyone stats page you can see the number of planes there are ADSB, MLAT, and other.
Here are the stats for the current day of Dxista:
498 planes without ADSB seen
393 of them were able to be MLATed
out of 2603 total planes seen
Europe has 10-20% without ADSB
US has 40-50% without ADSB
On the topic of MLAT. Is there a problem/conflict if two receivers close to each other (3 feet apart) have MLAT enabled on both?
The timing calculations have many factors. Basically the best timed sites are rotated for MLAT calculations. If there are 40 synced sites then about 10-12 sites are randomly picked.
The accuracy of a MLAT position depends on the timing of each site AND on the relative positions of each site. So 4 sites at the corner of a square will be much better for MLAT than 4 sites that are on the same line.
Here are your stats.
Understood, but what I was seeing gave me the impression that there was some kind of interaction between the two at the local level, that could be affecting the data as received, and interpreted, by Flightaware.
There might be a problem when you are in an area of 50+ synced sites. A site might not be picked in the random selection for MLAT. I have been seeing sites that have 80 locally synced sites so I am sure the totals are probably off a bit for individual sites.
There also might be some reason why the FA MLAT server picks one of your sites more than the other for MLAT.
I don’t find them useful after a couple of days. The issues they touched on are mostly covered in other more detailed and helpful posts. I’m a basic user, not an expert, the contents of the posts may not have been accurate or even correct in the end. No point keeping them, and confusing those reading them in the future.
Thank you for asking, and for the opportunity to comment on it.