STM32F103 with LoRa (I-CUBE-LRWAN)

[MChar.1]

I currently have a NUCLEO-F103RB board and also an I-NUCLEO-LRWAN1 board.

Is it possible to interface them directly together? I have seen that the I-NUCLEO-LRWAN is initialized for the STM32L0/L1/L4 series using the I-CUBE-LRWAN.

However, I would really like to interface it with an STM32F103. Is this possible? Can I use the I-CUBE-LRWAN software expansion for STM32Cube? If not, how can I do it?

Thanks

[MChar.1]

No one can reply to this message? 

Thank you in advance to the ST experts 🙂 !

[Tesla DeLorean]

I'm guessing most people would just port the software to there specific platform if necessary. Any reason to stick with the F1?

The board in question uses a USI module to do the LoRaWan comms, and does so via AT modem style commands, you don't have to program the radio chip directly. Use CubeMX or code in HAL some code to bring up a USART interface to talk to the module. Port in the AT comms code or code something simple to handle sending the command sequences and waiting for the responses.​ Review the datasheet and manauls for the USI module.

[MChar.1]

The reason for this is that we have a complex PCB based on an STM32F1 which already does a lot and is connected to different interfaces. The idea is not to change the MCU to add LoRa connectivity, but to add LoRa connectivity to the existing device. So we don't want to change the MCU just for that.

If I understand correctly, the USI module already includes an STM32L0. I can then use CubeMX, code in HAL to make a UART interface to communicate with the module. So I can use any STM32 (so my STM32F1) to communicate with the module with AT commands (in USART). So I can imagine using I-CUBE-LRWAN software expansion with my STM32F1 to communicate with the USI module. Is what I am saying correct? (not sure about that)

[Tesla DeLorean]

The Cube LRWAN stuff is mostly related to building radio level code, say if you wanted to replace the code in the USI module.​

Cube Expansion packages don't tend to integrate into Cube​ as well as the naming might imply.

Inspect the AT command project for the board, and port over the essence.​

[Tesla DeLorean]

One of the salient points here being that the examples use the SEMTECH hardware abstraction rather than ST's native HAL, so it is basically wrapped and forced into a different paradigm. I think some of that could be dispensed with if merged/ported sensibly.

STM32CubeExpansion_LRWAN_V1.1.1\Projects\Multi\Applications\LoRa\AT_Master (USE_I_NUCLEO_LRWAN1 fork)