STM32WL55 LoRa Stack issue

[IAnas.1]

Dear All,

I am using the STM32WL55 MCU running the LoRa stack in the dual core setup.

I am using STM32Cube FW_WL V1.1.0 to build my solution and upgrading to another version of the SDK is not a simple option.

The project is meant for the Australian region.

The issue we are encountering at the moment is that the device, when it sends data in  a certain range of bands it is mostly successful and when it sends outside of that list of bands it is about 50% successful

The configuration on The Things Industries that we are using is this:

Another point that might be interesting is that the device is sending a large amount of messages, so it could be that it is reaching a dutycycle restriction of some sort on the "good" channels and starts using the bad ones.

So my questions are:

Could there be some issue with the TTI configuration?
Could there be some configuration on the LoRa stack side that we can do to restrict the device in only using certain channels?

Thanks a lot in advance

[Issamos]

Hello @IAnas.1 

I suggest you to use the same configuration of the end_node exemple of the STM32_CUBE_WL .

Also, is always better to use the last firmware to avoid some firmware issues.

Best regards.

II

[IAnas.1]

Dear @Issamos 

I have some questions regarding the proper LoRaWAN standard that should be used on the TTI side and HYBRID_MODE.
In the code I see this:

And on TTI I see these options (with blue is the one I assume I should be using, but I am asking anyway):

Now my question regarding HYBRID_MODE and channels:
The frequency plan, to my understanding means that the gateway will be listening on channels 8 to 15. Is this correct?

From my understanding, during the join process the gateway dictates to the device which channels to use, right?

Is there any reason why would the end node choose to use channels that are not the ones that the gateway dictates?

Would HYBRID_MODE prevent this from happening?

Thanks a lot for your assistance

[IAnas.1]

@Issamos 

I realize that there is a contradiction between my original question where I am specifying that I am using the 1.0.2 specification.

At the moment the devices are indeed using that spec, but based on the finding in the code we should be using v1.0.3, correct?

In the scenario where 1.0.2 is used though is it normal that the device, after some time that uses the proper 8 channels starts using the other channels that are not being listened to by the gateway?