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.

IWR6843AOP: Adjusting Field of View for Static Detection

Part Number: IWR6843AOP
Other Parts Discussed in Thread: MMWAVEICBOOST,

Tool/software:

Hi there,

I’m working on a modified version of the Area Scanner demo and I'm trying to adjust the field of view for static detection.
Specifically, I attempted to modify the maxAngleDeg field in the heatmapGenCfg command to achieve a 60° FoV (-30° to +30°), as shown below:

heatmapGenCfg -1 1 0 2 62 30.0 3.0 10  


However, after making this change, the static detection functionality stopped working.
This leads me to wonder:

  1. Do the maxAngleDeg and angleStepDeg fields in heatmapGenCfg only act as filters, or do they also determine the heatmap size?
  2. If I want to adjust these fields properly, what factors or considerations do I need to account for to ensure static detection continues to function as expected?

Any guidance or recommendations would be greatly appreciated.
Thank you for your support!

Best regards,
Ed

  • Hello Ed,

    This is a good question. I am experienced with circumstances in which the main radar code will still work but the area scanner heatapgen portion and thus static detection doesn't. I have usually seen this happen when the heatmap configured to be generated is too large to be saved in memory. One thing you can test is lowering the max range bin the heatmapgen looks at to something small and see if it works. If it does, then yes it is because maxAngleDeg and angleStepDeg fields affect heatmap size.

    Best Regards,

    Pedrhom

  • Hello Pedrhom,

    Thank you for the suggestion, that's a good way to test for memory-related issues. I tried experimenting with different configurations to isolate the problem. Here’s what I observed:

    Configurations with maxAngleDeg < 50:

    heatmapGenCfg -1 1 0 2 10 10.0 3.0 10
    heatmapGenCfg -1 1 0 2 10 20.0 3.0 10
    heatmapGenCfg -1 1 0 2 10 30.0 3.0 10
    heatmapGenCfg -1 1 0 2 10 40.0 3.0 10

    In all these cases, static detection did not work.

    Configurations with maxAngleDeg >= 50:

    heatmapGenCfg -1 1 0 2 10 50.0 3.0 10
    heatmapGenCfg -1 1 0 2 62 50.0 3.0 10

    Here, static detection worked as expected, even with the full range.

    It seems the issue is not related to the memory size of the heatmap. Similarly, adjusting angleStepDeg also causes static detection to stop working.

    Do you have any idea why maxAngleDeg values below 50 or changes in angleStepDeg would cause static detection to fail? Is there perhaps an internal dependency or threshold for these parameters that the Area Scanner demo relies on for proper functioning?

    Thank you for your continued support!

    Best regards,
    Ed

  • Hello Ed,

    Generally due to the Area Scanner use case of sweeping a large room/field, such as an opening to an entryway, I have not personally experimented with low maxAngleDegrees values. If wanted to trim the area in which static detection is done, I would do so using the FOV parameters within staticDetectionCfg. Is there a specific reason you want to focus on the heatmap gen part?

    Best Regards,

    Pedrhom

  • Hello Pedrhom,

    Yes, because of memory limitations, the difference between maxRangeBin and minRangeBin is limited at 60bins, which limits the max distance I can use for static detection with a certain range resolution. If I could reduce the FoV processed in the heatmap, I could further increase maxRangeBin. The same goes for angleStepDeg I think.

    Best Regards,
    Ed

  • Hello Ed,

    Do you happen to have a MMWAVEICBOOST alongside the 6843AOP so that you can put breakpoints within Code Composer Studio (CCS)? I can point you to exactly where and when heatmap generation is done in the code.

    Best Regards,

    Pedrhom

  • Hello Pedrhom,

    That would be great, thank you. And yes, I do have a MMWAVEICBOOST alongside the IWR6843AOPEVM.

    Best Regards,
    Ed