The old matching process which tracked mode A values as pseudo-aircraft
got very, very expensive with a large number of mode A/C messages (and
with lots of single-bit errors, which seems common with a Beast doing
the reception)
Instead just count A/C messages directly into a 4096-entry array (which
is very fast) and periodically scan the mode S aircraft list to see if
we can match anything up (which is fixed overhead + cost proportional
to the number of mode S aircraft)
Reattach the licenses for the compat stuff to the source.
Only build/link the compat stuff when it's needed.
Rename compat/util.h so it's not confused with util.h.
Move all the platform specific defines inside compat/compat.h.
Notice synthetic mlat messages by looking for messages with a magic
timestamp value. If they arrive, tag the derived data as mlat-derived.
Don't include mlat-derived output in FATSV output to avoid loops.
This lets different things dynamically create the services they need,
and sorts out the horrible hacks that view1090 used to make outgoing
connections. Now you can explicitly create a service and tell it to make
an outgoing connection.
This means that view1090 can now just set all the ports to zero (to disable
the listeners), do a normal net init, then explicitly construct the beast
input service without a listener and tell it to make a connection as needed.
This gives access to the existing throttling behaviour when reading
from a file with --interactive, without needing to actually have
--interactive.
This is useful when testing, as without --throttle samples will be
processsed much faster than real-time, which can produce different
results with e.g. aircraft/filter expiry times and position/speed checks.
This currently understands:
UC8 (original rtl_sdr format, 8-bit unsigned complex);
SC16 (16-bit signed complex, full width);
SC16Q11 (bladeRF native format: 16-bit signed complex with 11 bits of magnitude)
Magnitude conversion now happens immediately when sample data is
received, so there is no risk of newly received data clobbering old
data under CPU overload.
This is possible now that the SBS output doesn't rely on the global block timestamp;
the output will look like this:
MSG,8,111,11111,4AC954,111111,2015/02/08,17:57:53.917,2015/02/08,17:57:53.936,,,,,,,,,,,,0
MSG,7,111,11111,392AEB,111111,2015/02/08,17:57:53.744,2015/02/08,17:57:53.936,,15375,,,,,,,,,,0
MSG,8,111,11111,392AEB,111111,2015/02/08,17:57:53.917,2015/02/08,17:57:53.936,,,,,,,,,,,,0
MSG,6,111,11111,800387,111111,2015/02/08,17:57:53.919,2015/02/08,17:57:53.936,,,,,,,,4745,0,0,0,0
where the "receive timestamp" (first time column) goes backwards to reflect the original reception
time of the delayed message, but the "forwarded timestamp" (second time column) reflects the actual
forwarding time.
the message being emitted immediately.
Fix computation of reception time so it's more sensible (the block timestamp
is some time after reception of the _end_ of the block, not the start) - this
means that message-emission times are always later than message-reception
times in SBS output, which is a bit more sensible.
Use clock_gettime in preference to ftime.
(except in --net-verbatim mode, where we emit them all)
Move aircraft tracking into track.[ch].
Clean up references to "interactive mode" when tracking
aircraft - we always track aircraft, even in non-interactive
mode.
since we have 8 bits spare, so there's no chance of confusing it
with an ICAO address, and we can safely use the filter table to match
future messages without also matching equivalent ICAO addresses.
Switch signalLevel back to a power measurement, don't put SNR in there.
But make it a 0.0 - 1.0 double so we're not scaling everywhere.
Adjust for the amplitude offset when calculating power.
Adapt everything else to the new scheme.
