ADR with RN2483 (The Things Node)

That assumes your gateway would forward all LoRa traffic that’s around. Given sync words and IQ inversion (and maybe even CRC) that might not be the case. Also, that does not take into account the transmission reach of your nodes?

An RTL-SDR dongle will tell you more, and is fun to play with. :slight_smile:


Im also working through this issue with LMIC,

I made an assumption (perhaps incorrectly) that TTN have adopted this ADR approach:

Semtech Lora networks Rate adaptation Class A/B specification

–LoRaWAN–simple rate adaptation recommended algorithm Common to Class A/B/C specification - Revision 1.0.

There are quite a few tweekable parameters…
Im also testing chirpstack which appears to use this same algorithm.

It would be very good to learn exactly what algorithm TTN are using and the parameters settings.

These LoraWan “grey areas” are when the fun starts!

The TTN sources are freely available on github so you can check for yourself…

Thanks Kersing!

Got it!


I comparison between the way ADR is implemented can kind off be seen by taking a diff between these two:

chirpstack-network-server/internal/adr/adr_test.go at master · brocaar/chirpstack-network-server · GitHub

ttn/core/networkserver/adr_test.go at develop · TheThingsArchive/ttn · GitHub

There are some interesting differences;