Newbie question about LoRa nodes (receive only?)

(Jcavera) #1

I've a project that I'm working on that requires long-range (10-ish km), mostly one-way communication at low data rates (100 baud is fine). Imagine a one-to-many remote switch. The transmitter will be centralized; but the receivers are many, scattered, and must be as small and low-power as possible. I've thought about doing a CW transmitter and stupid-simple RF detector for receive. But the problems with that is that (a) there could be other people on the band (915 MHz ISM) that I don't want to bump in to and a big honking CW transmitter would be a pain to them; and (b) that I may in the future want to add limited Tx capability to the other (currently receive-only) end as well. And so may as well build in that capability in the beginning.

So my question is this: Is LoRa the right thing for me to be looking at? Is there an easier way to do it? If I can get the receivers down to thumbnail size, I'll be happy.

So sorry if this has been covered before. I'm a true newbie to this and have only been thinking about it for a week or so. Any help or suggestions are appreciated. Thanks!

  • Jim

Edit: Just an FYI - I'm thinking NavSpark Mini and a HopeRF board to get the size that I need.

(Jcavera) #2

Yes, I'm replying to my own post. So a little more relevant information: The number of transmitters is exactly one, running at legal maximum (+30dBm) and in a central location. The number of receivers is (ideally) upwards of a thousand. The idea is that they need to get a "synch" signal at not-very-regular intervals (variable, but generally between 500 and 2000 msec.).


maybe this type of modules ? but these have nothing to do with TTN

but they are probably to big, maybe ANARDUINO's ?

(Jcavera) #5

I have a couple of Anarduino modules (actually HopeRF modules) on order, as well as a couple of Navsparks. I think that if I stack the one on the other and tuck a battery underneath, I can get the whole thing down to less than a cubic inch. Which would be cool.

I'm still unsure as to how the LoRa spec handles modules that are receive only. That bit is kinda confusing to me. Can anyone enlighten me on that point?

(Arjan) #6

So that one transmitter would send at least once per 2 seconds?

LoRaWAN (which is what The Things Network is implementing) is surely not an option for you:

  • LoRaWAN probably won't allow for sending that often, as even in best conditions (SF7) and when only sending a single byte you'd need to wait 4.6 seconds before sending the next packet on the same channel. And although LoRaWAN uses multiple channels, you probably won't be using SF7 to achieve a 10 km range?

  • LoRaWAN currently does not support broadcasting.

  • Even when you'd send a message to a single node: LoRaWAN nodes that are always listening are so-called class C devices, which TTN does not support, and as far as I know, neither do commercial providers.

Maybe bare LoRa would do, but I'd guess you'd also run into limitations of regulations. (If I understand correctly then LoRaWAN enforces duty cycles and rotating channels as it does not implement "listen before talk"; for bare LoRa there might be other ways to comply with your country's regulations.)

(Jcavera) #7

With my weekend reading, I had kinda suspected as much. Thank you for the confirmation, though. My fallback is to just use some bare sub-GHz receivers and "brute force" it, ignoring the needs of a particular protocol. Even though I won't be using LoRa, I'll still post results of the experimentation, as other may be interested to see how I get along with the HopeRF and NavSpark modules.

Thanks again for the help!

(Arjan) #8

Just to be sure: you could still use LoRa (the radio modulation), without LoRaWAN (the network protocol, which happens to use LoRa). Like the documentation of the Adafruit "Feather 32u4 with LoRa Radio Module" (which does not support LoRaWAN either, though it could probably be made to support that, using LMiC).

Of course, regulations apply (which in certified LoRaWAN modules are built-in).

(LoRaTracker) #10

You have not really provided enough detail on the project, in particular you say ‘10ish km’ but no information on the environment. This is critical information, the difference in expected range between an urban area and good line of sight is more than 1000:1

I do sell a number of PCBs that allow very small LoRa receiver\transmitters to be built, basically just a little bit bigger than an Arduino Pro Mini, you will find the details here;

A LoRa device is by its nature a transceiver, so with suitable code can be a transmitter and receiver.

You mention that there are upward of a thousand receivers, which is rather a large and expensive project for a ‘true newbie’, perhaps you should employ the services of a professional engineer, mistakes in the design of the project could be expensive.

(Jcavera) #11

That LoRa Tracker breakout is really cool – may need to play with it (though not for this project)! As far as the environment goes: Burning Man. So many square miles of absolutely flat nothing. For the receive/transmit part, you’re correct and I’ve determined that I’m not going to go with the LoRaWAN or even LoRa protocol, but just do a “non-protocol” with some HopeRF modules.

Yeah, an RF pro would be nice, but I’m a hardware hacker by nature and an engineer by profession. Though I’ve not done anything this large, 90% of the fun is figuring it out for myself! :slight_smile: