Please add mesh network ability

helium hotspots need a firmware update to allow them to operate as a mesh network. This would allow the deployment of helium nodes in locations without internet.

This would allow helium hotpot owners to expand the network coverage into regions where coverage is lacking.

Imagine being able to setup solar powered helium mesh nodes on key mountain tops around the country.

Helium coverage could then expand exponentially and truly compete with cellular coverage.

LoRaWAN is not (at least currently) compatible with Mesh.

You can’t do both at the same time, mainly because when the gateway is transmitting it cannot listen at the same time - so any packets trying to reach the gateway would fail. Also calculating dwell time for the gateway would get extremely complicated.

Exactly. In addition, mesh networks add a lot of overhead, which LoRaWAN was not exactly designed for.

A mesh network doesn’t have to be able to receive and transmit at the same time.

Seems people are doing just fine implementing LoRaWAN as a mesh network:

Interesting… I have messed with both Disaster Radio and Meshtastic, but I have not worked with a true LoRaWAN mesh. I am a big fan of OpenThread, however, and have implemented some protocol bridges with LoRaWAN there. Do you know of an open source LoRaWAN mesh implementation I could look at?

You can certainly build a Mesh network on top of raw LoRa - I was still at Pycom when development started around their OpenThread based LoRa Mesh implementation - or you can build a simplified version like the Nooropic Design you shared.

LoRaWAN however is a different beast, a gateway must be ready to issue downlinks at very specific time windows - and have the duty cycle to meet them - and you want to optimise the amount of time it is listening on air. So while you could build it the performance would be very poor. This is why I said it you cannot be operating both roles at the same time and expect to be good at either of them.

I actually doubt the Lemberg Solutions you shared is genuine as it’s marketing text very light on details on the mesh side and full of errors/misleading claims on the non-mesh parts of the text (“A single CR coin lithium battery enables a class C LoRa chipset to work for at least a year” - that must be some giant CR battery!)

What the LoRa Alliance has been working on for some time is an ability to use LoRa as the gateway backhaul, but this will very likely necessitate different hardware (separate TX and RX antennas for example)

1 Like