This is just some information that I have found.
While playing with basic lora and 4 nodes, all close to each other, I was getting a higher then expect miss on packets not arriving.
Note this is from a single TX, no retries and just my low level lora testing.
I setup my spectrum analyzer to have a look and I can see a fair amount of radio activity in the lora AU915 band. I turned off all my lora devices and the back ground signals remained.
The signals are fairly strong (for lora) at about -80dbm at my workstation.
After a bit of googling I discovered that in Victoria the smart meters use a few bands including the 915Mhz to create its mesh to get the smart meter readings and send commands out to the smart meters.
So I took my analyzer out to the meter box and confirmed that in my area the smart meter seems to be using the 915Mhz band, thus competing with lora/ meshtastic etc.
i.e. The signals got stronger.
The usage is constant, but nowhere near 100% duty cycle, so there should be room for other lora packets to get transmitted, but there seems to be a reasonable chance that they can interfere with each other.
My key take away from this is to ensure I acknowledge a packet and that the transmitter should retry a few times if it does not hear the ack.
This also means you need to manage duplicate packets as the node tx may work, but the ack may not get back.
Note: this information is provided as is and I note that this is in my area and others my not have the same challenge. But it you are having some issues with lora in general were it seems a bit hit n miss, it may be due to band utilization.
Happy to discuss or for others to add any interesting findings here for discussion.
