• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 54 subscribers
  • Views 1964 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

CC1310: 15.4 Stack FH mode Orphan State behaviour

Abrar Magdon
Intellectual 585 points
Part Number: CC1310

I am trying to understand the behavior of the 15.4 Stack Frequency Hopping mode ORPHAN STATE.

I did a to test to see how the sensor behaves when a collector powers down or if the sensor goes out of range from the collector.

What I noticed when I powered down the collector for about 5 mins and turned it on again, the sensor at times rejoins the collector but at times it does not rejoin the collector (checked over 1 hour).

By repeating the process it it possible to get the sensor to rejoin.

I also tried this with multiple sensors. At times all sensors rejoin and at times only some rejoin.

The results are not consistent.

I would like to understand the behavior of the stack program in Orphan State.

Could someone please explain what the sensor is suppose to do when it is in orphan state, and why it does not rejoin all the time?

Thanks in advance.

  • 0
    TI__Expert 4375 points
    Hey Abrar,

    I am not noticing the sensor device not being able to rejoin the fh network. I am using the stock out-of-box sensor and collector hex files found in the SimpleLink CC13x0 SDK v1.40
  • 0 in reply to Brocklobsta
    Intellectual 585 points

    Hi Brocklobsta,

    I am using the SimpleLink CC13x0 SDK v1.30 example project (NOT hex files).

    I used 4 CC1310 Launchpads as 4 sensors and another Launchpad as a collector. 

    The frequencies are set to 915 MHz, the Pan ID and net names are changed, Frequency hopping is enabled, security is disabled and the reporting interval is 5 seconds.

    The example project with these modifications are compiled and loaded to the Launch pads.

    I also have another setup with 10 custom boards as sensors with the above modifications and a custom application running in the same program. A CC1310 Evaluation module is used as a collector.

    In both these setups I am facing this issue. All sensors do NOT reconnect All the time, to the collector after they become Orphan. Its not reliable. I have tried keeping them even for more than an hour.

    I tried both, powering down the collector and powering it up again, and also, taking the sensors far from the collector to loose signal. 

    At the moment I am using a software reset after 15 minutes as a temporary solution if they fail to reconnect to bring them to the Joining/Rejoining state.

    What could be causing this issue?

    Could this be because there are many sensors trying to reconnect at the same time?

    Additional question: After a while after the sensor becomes an Orphan, the collector indicates !Responding for the specific sensor. Is there a difference in the way the collector behaves before and after sending the !Responding message. 

    Thanx in advance,

    -Abrar

  • 0 in reply to Abrar Magdon
    Intellectual 585 points
    Hi Brocklobsta,

    I tested this issue again using 4 Launchpads as 4 sensors and another Launchpad as a collector.

    I have used the CC13x0 SDK v1.30 example project and changed the reporting interval to 5 seconds, Enabled FH, disabled Security and changed the Pan ID / Net name.

    I let all 4 sensors join the collector and left them to work for a while. Then I took the sensors far away from the collector until they loose connection. The sensors Indicated state changed to Orphan and the collector indicated !responding for the 4 sensors.

    I left them in Orphan state for 25 minutes. Then I brought them back close to the collector for them to rejoin.

    Sensor_4 rejoined as I came close to the collector
    Sensor_3 rejoined at 12 minutes
    Sensor_2 rejoined at 48 minutes &
    Sensor_1 rejoined at 50 minutes

    Why do they take so long to connect. At times I have left them for over 1 hour and still some sensors do not rejoin.


    Any advice would be much appreciated.

    Thank you
    -Abrar
  • 0 in reply to Abrar Magdon
    TI__Expert 4375 points
    Abrar, the collector outputs !responding after unsuccessful tracking. The behavior should not change except that a tracking request will be sent out when hears the device again. I recommend upgrading to v1.40 to have the latest bug-fixes.
  • 0 in reply to Brocklobsta
    Intellectual 585 points
    Hi Brocklobsta,

    I tested this issue again with the Stack SDK v1.40 but the behavior was the same.

    I opened the collector and sensor projects.

    In both config.h files changed

    /*! Security Enable - set to true to turn on security */
    #define CONFIG_SECURE false
    /*! PAN ID */
    #define CONFIG_PAN_ID 0xD003
    /*! FH disabled as default */
    #define CONFIG_FH_ENABLE true

    In the collector.c file changed

    /* Default configuration reporting interval, in milliseconds */
    #define CONFIG_REPORTING_INTERVAL 5000 //90000

    Programmed 4 sensors in 4 launch pads and a collector in another launch pad

    once 4 collectors joined and was transmitting data to the collector I took the sensors away from the collector to loose signal.
    Stayed there for about 5 minutes and brought them back to the collector.

    Within about 1 minute Sensor_4 rejoined and started transmitting data to the collector.
    After around 30 minutes Sensor_3 rejoined and started transmitting data to the collector.
    I have let them run for 2 hours now. Still Sensor_1 and Sensor_ has NOT rejoined yet.

    What I have noticed from all the tests I done is that 1 sensor rejoins immediately and others take a long time. Some don't connect at all.

    Could you please check this from your end and confirm if you see the same behavior.

    Thanks in advance.

    -Abrar