Sure they do. Or rather they can’t figure out how to pack multiple 10 bit datagrams into 1500 bit packets.
https://opensky-network.org/community/publications
I am pretty sure they would be able to figure out
That’s not what’s happening, they just process the beast stream server side, piaware processes it on the pi.
Seems like they just tore a page out of @prog and @wiedehopf book with the recent airspy commits: Novel error correction algorithms for ADS-B signals with matched filter based decoding - ScienceDirect
Be curious to see the entire article.
I think this should be it: https://ieeexplore.ieee.org/document/8767962
Edit - actually this may be a different article, but it seems to cover the same research.
Edit 2 - I found a copy - it actually describes a different algorithm to the one I linked above. I won’t post it directly here, since it’s not exactly legit but if you want to see it, it’s easy to find it on sci-hub.
Created by three people from a country where ADS-B sharing to outside is not permitted . funny fact
In the same country there is also no law to protect intellectual propriety and sending “visiting professors” to gather “information” is a state sponsored business. Just another fun fact.
The approach is similar in some aspects like the matched filter, but the SNR based FEC was not implemented.