I am working as a Embedded Systems Intern in a company so beware of my ignorance.
My question is ;
I was thinking that we are just working with a LoRa module and a MCU, that entegration was needy. Yesterday, i realise that this is wrong. Because, some of my seniors said to me that we can burn a custom firmware for a module and work only with it. We dont need a MCU; because it has already a MCU in a module.
Just please can somebody explain me every part of that step? Assume that we are using a STM32 based LoRa Module. How can we burn a custom firmware for it?
It will depend on exactly which modules, what ports, what IDE is supported, what coer LoRWAN stack embedded etc. Can you provide more details then some of the more experienced guys may step up to offer guidance and/or refer to prior TTN labs and documentation, youtube vids, Forum search or whatever. Note most likely though, I suspect many will guide you away from just crashing in and mod’ing firmware on sole MCU, to run you application in parallel, unless you fully understand LoRaWAN specs and eg. The often tight timing requirements involved in getting custom code working with a reliable LoRaWAN implementation esp. in the context of latest TTN (CE) V3 instantiation…
ST’s fork of Semtech’s reference LoRaMAC-Node supports their own radio + MCU modules and has available patches from makers of others, eg, Murata.
The simplest solution there is not to try to do anything in between transmission and the following receive windows.
After that, a typical Class A node is basically free to do whatever it wants timing wise until it next transmits, provided that the stack is aware of the passage of any time spent asleep so that frequency utilization vs. time tracking comes out right.
esp. in the context of latest TTN (CE) V3 instantiation…