BLE to LoRaWAN "relay"

Hi everyone,

I hope anybody has a solution that is certified already and can be integrated in our LoRaWAN-eco-system. So it should be platform/LNS independent and not requiring a manufacturer proprietary server/application.

I´m looking for a LoRaWAN and BLE device that works as a BLE to LoRaWAN relay for iBeacons,
conbee GmbH was developing such a solution and I had working prototypes of the L300 (https://conbee.eu/de/loesungen/#products) but these devices are not sold anymore. Don´t know what happend, but seems like they are going out of business…

So what I need is a water resistent, quite small BLE to LoRaWAN device, that can be configured to scan for specific iBeacon IDs and sens the device MAC, signal strength and perhaps battery status to our LoRaWAN.

Is there anything available. Couldn´t find ready solutions. Only Dev boards…

Thx and keep yourself safe,
Robert

Ok, found Hardware offered by LANSITEC and hope it works.

Translation tasks are often very hard to define when semantics mismatch.

BLE tends to be very repetitive and chatty, but LoRaWAN needs to be infrequent and terse. Even a full BLE UUID is really a bit long for LoRaWAN.

This all means that a good solution to the problem is likely to require an algorithm expressed in custom code, not as configuration knobs for a stock algorithm.

You’d want to do things like restrict the patterns scanned and report them in some more concise format. You’d want to take a set of received signal strength reports and drastically condense them to a single summary number weighted in a way that make sense for what you are using it for. And maybe you’d want to report appearance/disappearance, through the lens of some strategy for deciding what sort of gap in receptions indicates disappearance…

And you also have to remember that LoRaWAN is not reliable - if you condense your report and it doesn’t get through, how do you handle that. Do not attempt to use confirmed uplinks - they’d don’t work the way you might imagine (pretty much, they don’t work - downlinks are already expensive, if you must do them, do them on a long term strategy at application level)

In other words, you probably want off-the-shelf hardware on which you can run your own code, rather than a “product”.

1 Like