I am using the SX1272 in and XBee module from Libellium and the SX1276 (built into an RN2483 module from Microchip). Most of my experience is with the RN2483.
Well, you can’t run netmf ‘on’ those boards, but I am driving them from netmf. The driver code is code that I wrote. It supports Lora, but not full LoraWan yet (because the gateways are too darn expensive). I can share the code for select non-profit uses or will license it for commercial apps. Works on netmf, Win10 Iot Core, Win10 and Mono.
So, I received these boards today and I’m thinking about producing a Gadgeteer breakout for them (with the passives and an antenna connection). Price would be around $30-45 each I’m guessing. Considerably cheaper than the Libellium XBee Lora plus the XBee adapter.
Questions:
Is this interesting?
Would you need North America (433/915) or Euro/Asia (433/868 ) frequencies?
U.FL or SMA antenna connection?
Should I try to offer a PCB inverted-F antenna option (no external antenna required, but with a reduced range)?
Then end result would be a Gadgeteer radio network with 3-22km point-to-point range, and compatibility with GHI Raspberry Pi hats for use as a controller or a gateway device to other networks.
And no, these would not be CE or FCC certified, and would (at least initially) only support Lora, and not LoraWan (which requires a gateway device and consortium license).
EDIT: Yeah, but I started out as a hardware guy mumble-mumble years ago.
So yes, it is straightforward to build a working gateway of some sort - it’s building LoraWan compatible gateways (which require licensing of some sort from the Lora consortium) that are expensive and which I am not promising to interop with yet. LoraWan gateways use a very specific protocol to tunnel lora-to-lora messages through the internet or other non-Lora backbone.
So, I’m only shooting for working, long-distance, Lora-to-Lora interop at this point, and you can create your own gateways if needed and like @ .Peter. illustrates.
Yes we will certainly want an antenna, but the tiny ufl connector is not meant for the prototyping process, where you pull your board out from under your notebook or coffee cup and plug the thing in, use it and disconnect it again. It is meant for plugging it in closing the case and shipping it.
Note the people (including me) having trouble with their antennas on the new Particle Electron.
I vote fort the SMA connector. If we are going for range, we might as well get as much of it as we can.
On the other hand, some applications may not need the super range and no external antenna might be nice.
Can someone with a brain, point out the differences between 433 wireless and 433 LoRa? Is it just that LoRa is more conservative on the power usage and has longer range?
And why would we want a LoRaWAN gateway if we are just receiving data and sending it off to the Azure storage depot? I have read, but I am not comprehending.
Thank you for any light you can shine in my dark object on my shoulders.
If I can swing it, I’ll leave pads for both antenna fixtures, and just populate one of them - that way we can support both.
Lora is just a particular modulation and protocol. Other radios use the same freq in other ways.
You don’t need LoraWan for Lora-node to Azure. You need LoraWan for Lora node to internet to Lora node. The gateways use a specific protocol to ensure that they can transparently route packets between nodes that are out of range of each other, using the internet backbone.
So your lora node attached to a ghi board sending data to a raspberry with a lora node or any board with access to Ethernet or wifi as gateway to Azure. Correct?