Add LoRa middleware/driver to different boards to use CubeMX

[Jonas Chr]

Posted on June 13, 2017 at 11:05

Hello folks,

it seems like there is a very important topic not discussed right now. There are LoRa examples for System Workbench for 5 different Boards, which include in the folder middleware all functions of LoRaWAN. But as the code of drivers and middlelware is not easy to understand, it would be nice to migrate it to another project on another board. I just want to pass over the data to that middleware, so I don't have to worry about the further handling. And by the way the modularity, which STM32 describes on nearly every page, should work somehow!

My working steps so far:

I found Cube (without MX), but there is no driver/middleware/libraray for LoRa.

Also tried just to copy the Folders Driver and Middleware of the STM32CubeExpansion_LRWAN_V1.1.0 to a new CubeMX project. Would be awsome, if there is a tutorial, how the migration of drivers and middelware is realized 🙂

Thanks in advance

Jonas

#middelware #lora #lorawan #cubemx #driver #migration

[Tesla DeLorean]

Posted on June 13, 2017 at 15:10

Probably not getting a lot of discussion because the people who are working on it and understanding it have their own products to work on? The problem with your approach is that it requires other people with more insight to have to explain everything in more detail than they needed. Generating teaching materials for the second or third wave is a trailing priority. Being in the lead group requires you to dig into the material already available/provided.

If you are uncomfortable with the protocol or the stack, then you should do some research on those. The stack itself is from Semtech so should be consistent with other implementations on other platforms.

Work with the code on the platforms it ships with, get comfortable with what ties it to specific resources and pins. Then find the platform closest to your target and port that. If you need assistance from an FAE or contractor consider those options.

[Amel NASRI]

Posted on June 13, 2017 at 16:53

Hi

christofzik.jonas

‌,

I would like to clarify the following:

1- The expansion packages (as

STM32CubeExpansion_LRWAN) aren't supported in STM32CubeMX

2- For each Cube package, there is a set of documents available to get initiated with package usage. The package contains drivers as well as simple and complex examples for all available boards & peripherals.

3- Depending on the product you will be using, you have to select the STM32Cube package. Documentation related to STM32CubeF4 for example is available from

http://www.st.com/content/st_com/en/products/embedded-software/mcus-embedded-software/stm32-embedded-software/stm32cube-embedded-software/stm32cubef4.html

.

4- What is the other board that you would like to use? Depending on the selected board, some HW modifications may be required for your board or the Lora shield before making the FW updates.

-Amel

[Tesla DeLorean]

Posted on June 13, 2017 at 17:49

Microsoft had this same problem calling everything .NET when it meant different things in different contexts. 

Branding that confuses people is of negative value.

You really need to stop calling them Cube Expansions when they are HAL Examples. 

[Tesla DeLorean]

Posted on June 13, 2017 at 18:00

>>

4- What is the other board that you would like to use? Depending on the selected board, some HW modifications may be required for your board or the Lora shield before making the FW updates.

The Shield board could do with build examples for F4 and F7 DISCO boards, among others, coverage of all STM32 family parts across all boards you have Arduino shield connectors on. The focus currently is entirely on 'Low Power' which makes sense for end-node products, ignores a much larger universe of systems where power is not the primary constraining factor. Consider situations where power is available, but installation/repair costs are very high, you don't want to service things with batteries that fail.

[Jonas Chr]

Posted on June 14, 2017 at 09:03

Hi

st.mcu

‌

thank you for your reply, would be very nice, if you help me with that topic.

1- Good to know, that it is not my fault 😉

2- I guess every example in Cube, could be build much easier and quicker in CubeMX. The point is, that therecould be a Folder Lora under STM32Cube_...-> Middleware -> Third_Party, like it is implemented in the example. So I can not find something for LoRa peripherals.

3-I did find out, how to import from the examples HAL, I2C, SPI and so on in the GettingStarted. The next steps could also be mentioned there, so that you don't need to look in other applications/examples, how to use the imported project. For beginners that just would be a nice to have.

4- We want to go with the Nucleo L452RE, to set up a LoRa Node. I guess in addition to the HAL_Driver we just need a LoRa library to start with that.

Kind regards

Jonas