SoftRF version MB171

[Moshe Braner]

https://github.com/moshe-braner/SoftRF

Changes between MB167 and MB171:

Enhancements:

Allow use of the frequency correction setting on sx1262 too.

On T-Beams, after booting, show the main configuration (ID, protocol, aircraft type) OLED screen first, for 8 seconds, then auto-switch to the TX/RX screen.

Revised the air-relay modes "landed out" and "ADS-B too".  The latter is for relaying ADS-B traffic by T-Beams that have the ADS-B receiver module added.  This spreads the benefit of ADS-B traffic awareness to other nearby SoftRF devices that do not have the ADS-B module.  This is transmitted in the "old protocol" thus remains invisible to FLARMs.  (PowerFLARMs have their own ADS-B receiver built in.)  Landed-out traffic (SoftRF that has been manually switched into landed-out mode) is relayed in the "new protocol", the idea is to make such situations as visible as possible to other SoftRF, FLARMs, and ground stations.

Revised the "relay only" mode.  It used to be for relaying FLARM traffic, which was not actually usable since FLARMs react badly to seeing their own ID in incoming packets.  Now this mode is for local relaying of ADS-B traffic by T-Beams that have the ADS-B receiver module added.  This is intended as a way of adding ADS-B traffic awareness to systems based on Classic FLARMs, wirelessly and without any changes to the existing avionics such as FLARM viewers.  It can also spread the benefit to other nearby Classic FLARMs.  This mode is not useful in conjunction with PowerFLARMs, since they have their own ADS-B receiver built in.  These transmissions are done in FLARM-compatible protocol, at low power.  Own-ship position is not transmitted in this mode, but reception is normal, thus can also feed data (via WiFi or BT) to XCsoar at the same time.  (Set to same ID as the FLARM, then SoftRF will ignore the transmissions from the Classic FLARM.)

Bug fixes:

Revised the handling of GNSS dropouts & leap seconds.

Fixed preamble & sync word on sx1262 (mainly: T-Echo) so it can be received more reliably by OGN ground stations and the latest (multi-protocol) OGN trackers.

Fixed a bug that delayed first transmission to a full minute after a GNSS fix is obtained.  Now the delay is 30 seconds, intentionally to better ensure the stability of the fix.

Fixed some bugs in the air-relay code, and tested to make sure the relayed data is accepted by both PowerFLARMs and Classic FLARMs.

Introduced a delayed-relay method to allow much more reliable relaying (usually every 7-9 seconds in the "ADS-B too" mode) of ADS-B traffic.  This was necessary since most ADS-B position packets seem to arrive in the first half of each second.  Thus most of the ADS-B packets arrive after own-ship transmission already happened (disabling further transmission during the time slot), or during the dead period, 200-400 ms after PPS, which is not part of the FLARM time slots.  Now the relay is done later in the second time slot (400-800 ms after PPS), pre-empting own-ship reporting (which was already done in the first time slot).

Removed the pre-transmission sanity check limiting vertical speed to 20 m/s if relaying a jet - apparently airliners climb faster than that!

Happy New Year everybody!

[Moshe Braner]

Needless to say, I can't do all this by myself.  Several people helped with these issues, making suggestions, reporting bugs, and spending a lot of time testing things.  You know who you are.  Thank you!