How to improve my LORA-based project?

[Jean-François Milants]

Hello,

I've created a simple project : I want to measure my water consumption. The water meter provides me with a pulse output, so it's easy to integrate with an MCU (a LOPY board, in my case).

As the water meter is out of reach of my wifi network, I wrote a small micropython script that counts the pulses and sends the total to another LOPY board using node to node raw Lora (based on this example : https://docs.pycom.io/tutorials/networks/lora/module-module/).

This 2nd Lopy forwards the message on MQTT via Wifi. The data is then processed by Home-assistant.

This prototype works as expected. I'm in the process on improving the proto (waterproof enclosure, battery, sleep mode,...).

My question here is about the Lora communication. For now, the sensor board just sends the number of pulses using raw lora (no protocole, formatting, encryption,...) and the receiver board just listens to lora and parses the messages it receives.
Is there a better way to do this? An existing protocole I could implement? Let's say I would like to send more data (temperature, uptime, battery level) and add more nodes, should I just improve my node 2 node protocole, or should I use a Lora Gateway? Join a Lora network like TTN?

What's the state of the art for a small network of Lora sensors?

Thanks!