This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CC1352R: TI15.4 Stack - FH Max Broadcast Interval

Part Number: CC1352R
Other Parts Discussed in Thread: SYSCONFIG

Hey everyone,

I have been testing the broadcast interval in FH mode for SDK 7.10. We need the interval to be equal or larger than 5 min. However, nodes don't seem to receive the broadcasts when the interval is larger than around 2min.

If the broadcast interval on sysconfig is defined as 2min it works fine, but if defined with 3min or bigger, nodes dont receive them. 

The stack user guide says that the broadcast interval can be from 15 to 16777215 ms. However there seems to be some hidden limitation.

Can you help?

Thanks.

  • Hi,

    Can you give some more details on your configurations on sysconfig for the sensor and collector?

    Regards,

    Marvin

  • Hi Marvin, default TI Sensor Collector demo, FH mode, all channels, 868MHz. Only changed broadcast interval.

  • I will do some tests to check if I get the same results. Could you in the mean time try and use a sniffer to see if there are any packets being sent?

    Regards,

    Marvin

  • Hi,

    I have done some tests and the nodes are still able to receive packets from the collector even with 3min broadcast interval. Checked both with the sniffer and LNA pin of the nodes. Do you have any logs from your tests and have you been able to use the sniffer?

    Regards,

    Marvin

  • Hi Marvin,

    How can you be sure that the sensor received the broadcast message by those methods?

    - Sniffer only shows that the broadcast was transmitted (which it was every time, I never said otherwise) but there isn't any broadcast ACK so you can never be sure the node received it.

    - The LNA also only proves that the RX was ON but it doesn't guarantee it was correctly received and processed. Also might be hard to identify LNA pulses related to broadcasts, as sometimes reports and polls coincide and create similar ON pulse durations.

    The method I used is just observing the red LED on the sensor node. If the network is left alone, the red LED on the sensor is turned ON when joining, and it must toggle every time it receives a broadcast message. If I choose 10s broadcast interval, after 60s from startup, then the red LED on the sensor toggles every 10s which shows broadcasts are working well. If I change the interval to 180s or 3min, after the initial 60s startup, the red LED never toggles again which shows broadcasts are not working.

    Can you perform this test instead? I don't think it makes sense to send you a 3min video showing no activity on the red LED.

    Thanks

  • Hi again,

    I was looking at the unicast data exchange, which is not the same as the broadcast exchange. My mistake. I can see the issue. I seem to be getting it when I increase the broadcast interval over 130sek. I am using a logical analyzer to see the data sent from the collector (PA) and the data received by the node (LNA).  I am toggling a gpio every time generateBroadcastCmd() function is executed in the collector, since that is when the collector sends broadcast request to the mac. I am also toggling gpio when processBroadcastCtrlMsg() is executed in the sensor, which is when the node receives the broadcast packet. I increased the broadcast packet length to 512 to be able to differentiate the broadcast packet and the other data packets. The collector is able to still send the broadcast packet, however, the node is not receiving it for some reason. This is also reflected with the led not toggling as you mentioned.

    I will try and look into what the issue could be and get back to you with an answer. 

    Regards,

    Marvin

  • Hi Marvin,

    Thanks for checking again. Yes that is the same conclusion I got, over 130s it stops working 

    Please do look into it, I really need longer broadcast intervals for my application. Hopefully there is an easy fix somewhere.

    Looking forward to your feedback.

    Thanks 

  • Hello Marvin, any update on this matter? 

  • Hi, 

    Yes, I found the issue and will coordinate with R&D to make a fix for it. The bug was in the mac code, so we don't have a workaround for making your project work immediately. It will be fixed in the next SDK version. I apologize for the inconvenience. 

    Regards,

    Marvin

  • Hello,

    It's unfortunate that there isn't a workaround, but at least it will be solved in the future.

    Thanks.