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.

cc2538 end device association problem

Other Parts Discussed in Thread: CC2538, CC2592

Hi TI,

Below is the test I have performed using cc2538+cc2592 ZigBee module with Z-STACK-HOME:1.2.2a Loaded in it.

1. Started COORDINATOR-1 with 0xABCD PAN id by allowing it form a network in best channel by scanning all the channels. As a result Coordinator started in 0x14 channel

2. Started a END DEVICE with 0xABCD PAN id by allowing it to scan all  the channels. As a result End device has joined to the above coordinator.

3. Started another COORDINATOR - 2  with 0xABCD PAN id by allowing it to scan all the channels . As a result Coordinator has started in channel 0x0B.

4. Turned off the COORDINATOR-1 which is in channel 0x14.

Observations:

1.The end device which has lost its parent releases orphan notification and scans all channels for a parent.

2.COORDINATOR - 2 responds to the beacon request of the end device .

But the end device does not get associated with the COORDINATOR - 2 . Why is this happening?

The end device is continuously scanning all the channels without any successful association with a FFD.

  • Do you intend to clone a ZC? It doesn't work that you assign the same PANID to two different ZC.
  • A network can only have one coordinator.
    In your system, you are setting up two coordinators with the same PANID and non-overlapping channel mask, while your end device has probably both channels in its channel mask.
    That is probably why the second coordinator, when brought up, doesn't see an existing PANID already up and running and therefore creates a network with the same PANID, while the end device scans for both channels, right?

    The uniqueness of the network is however represented by the extended PANID. When the device orphans from the first network to the second, he sees the network with the same PANID but with a different extended PANID in the beacon message.
    That is why he discards the second coordinator.
    Not sure what you are trying to achieve with the second coordinator but the behavior of the end device is normal.
    TheDarkSide
  • The uniqueness of the network is however represented by the extended PANID. When the device orphans from the first network to the second, he sees the network with the same PANID but with a different extended PANID in the beacon message. That is why he discards the second coordinator.

    If that is the case I see that when I start a new router while the end device is scanning all the channels , the router gets added to the powered on coordinator and then immediately the end device gets added to the router. I clearly see that the end device gets added only when there is a new device joining or starting a network(either starting a new coordinator in any channel and with any PAN or starting a new router with presence of a coordinator ).

    Hope am clear.  

    Am trying to understand the network behavior under different cases, for better understanding of the stack.

  • If you start a new router while the end device is scanning all the channels , the router gets added to the powered on coordinator. But, the end device won't add to the router.
  • The router gets added to the powered on coordinator as is expected , and then the end device joins the router as I could see. I will upload the sniffer file by performing this test now again.
    I would make the ED orphan and start a router with an already existing coordinator in some other channel.
  • Do you enable NV_RESTORE in your test and does device do power reset?
  • Hi Yikai Chen,
    I was wrong in saying that the end device gets added to the newly added router. I have tested it now. The Orphan end device gets added to the newly started coordinator(the coordinator which I have powered off initially).Because the extended PANID that the end device has will be same as that of the newly started coordinator.

    Thanks for the clarification
  • You are welcome and hope you have a clear picture now.