• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 366 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

IWR6843ISK-ODS: To improve cfg parameters

Smart Fountain
Intellectual 320 points
Part Number: IWR6843ISK-ODS

Hi,

To evaluate people tracking/counting performance in our environment, I have changed the parameters in cfg file.

The reason of the change is there are restrictions in our environment as following,

- Bandwidth shall be smaller than 0.5GHz.

- Duty cycle shall be smaller than 7%


The test have done using following demo.

\ti\mmwave_industrial_toolbox_3_4_0\labs\lab0020_pplcount_Overhead\GUI_A_base_no_logic


Although I had prepared that range accuracy would be poor because bandwidth became narrow, I found some more issues.

- There are only few points associated to the person.

- There are many points associated to the ID 255.

Would you tell me which parameters should be change to achieve better performance ?

Please find attached the cfg file.

Best regards,

Fountain

  • 0
    TI__Mastermind 28055 points

    Hi Fountain,

    Limiting the bandwidth to <500 MHz will limit your range resolution. This will cause you to get less point per person, all other things being equal, and cannot be avoided. 

    Points associated with 255 are points that don't belong to a track.  If you are seeing a lot of these points around a tracked target, increase the gatingparams. If these '255' points are separate from tracked people, then it is working as intended.

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Intellectual 320 points

    Hi Justin,

    Thank you for your comment. May I ask you a few questions ?

    1. Do you have any concern in our current cfg setting in terms of the chirp signal shape to adopt our restriction (i.e. 0.5GHz / Duty < 7%) ?

    The intention of this question is to verify validity of our parameter changing.

    2. What do you think about changing CFAR parameters (ex. decrease threshold) to increase points (though the change would increase unnecessary points...) ?

    3. The gatingparams are 3 2 2 2 12 in the current cfg file. They are thought to be large enough, aren't they ?

    Best regards,

    Fountain

  • 0 in reply to Smart Fountain
    TI__Mastermind 28055 points

    Hi Fountain,

    1. Please ensure that your chirp is in the allowed band, if applicable. So if the allowed band is 61-61.5, ensure that you are starting at the right frequency.

    2. You will need to experiment with this yourself - with high doppler resolution you may still be able to get a good number of points per target. If you are getting a lot of random points, you only need to worry if the tracker is behaving poorly because of them.

    3. I believe those gating parameters are good. You can also try: 3 1.5 1.5 1.5 12 for a more precise gating function.

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Intellectual 320 points

    Hi, Justin

    1. It is allowed to use 500 MHz within 60-61 GHz in our environment. So the start frequency was set to 60 GHz.

    2. I will try some parameter changes. Thanks.

    3. You have told me that " If you are seeing a lot of these points around a tracked target, increase the gatingparams.", In accordance with that advise, the precise gating parameter would work opposite result, wouldn't it ? Do you have any recommend values with the parameters to increase points ?

    Best regards,

    Fountain

  • 0 in reply to Smart Fountain
    TI__Mastermind 28055 points

    Hi Fountain,

    To increase points, lower the CFAR thresholds.

    cfarCfg -1 0 0 8 4 4 0 6120
    cfarCfg -1 1 0 4 2 3 0 6120


    These are the bolded arguments in the above cfg commands.

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Intellectual 320 points

    Hi, Justin

    Thank you for your advice.

    Yes, I'm going to try several values about Threshold in cfarCfg.

    On the other hand,  I would like to hear about gatingparams, because it is your first proposal. Should it be tried them as well?

    Best regards,

    Imaizumi

  • 0 in reply to Smart Fountain
    TI__Mastermind 28055 points

    Hi Fountain,

    Gating params are : <Gating Gain> <X Limit> <Y Limit> <Z Limit> <Doppler Limit>

    Each track has a dimension determined in X, Y, Z, Doppler Space.  The dimension is multiplied by the <gating gain> to find points to potentially associate with the tracked object. These dimensions are restricted by the Limits (Last 4 arguments).  Setting a limit as zero removes the limit.

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Intellectual 320 points

    Hi Justin

    Thank you for your reply.

    I will try changing gating parameters referring to your explanation.

    Best regards,

    Fountain