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.

WEBENCH® Tools/IWR1443BOOST: To solve ghost detected point problem.

Part Number: IWR1443BOOST

Tool/software: WEBENCH® Design Tools

Could you advise us to solve ghost problem?

we found some ghost noise when nothing in front of the sensor field of view.

We mean "ghost" that is detected points on the view in spite of there is nothing on the floor.

How should we do to solve (or clear) this ghost?

Shold we re-cariblation? or Shold we change setting?

cf. attached file is test environment and settings when the ghost happens.


Regards,

Senoo.

  • Hello,

    In the picture of your EVM setup (hardware), If we have a height above the floor of 70cm, and you imagine shining a flashlight with a cutout that matches the azimuth and elevation Field of View,   the radar reflection is the difference frequency, where the receiver signal level is above a threshold of a detected event.

    a) if you are 70cm above the floor, there is a ground reflection that can return.  If there is any mechanical mounting like the baseplate in front of the radar antenna EVM, it can be a reflected object.

        - remount the radar using the 4 screws from a mounting plate behind the radar, and make sure the baseplate is behind th EVM,

        - this can also happen if there is some foreign material on the EVM antenna.   Carefully using an filtered air spray, and spray the antenna area

        - changing the elevation angle of the EVM antenna relative to the floor, can also help to remove the stand or ground relflection. 

         - sometimes thinck corregated cardboard can act as a radar deflector laid at the base of the EVM floor mount.

    b) in your picture of the receiver returned signal (Range Profile for Zero Doppler)

          after you have remounted the radar, and cleaned the antenna

           there is a strong signal at DC, this may require saving the Visualizer config to PC, using the mmwave SDK User Guide, adjust the  calibDcRangeSig API

           values, to minimize the DC (0 meter) amplitude

            as the DC level is reduced, using the same graph, the detection (orange dots), should eventually be removed near 0 meters.

            There is also a visualizer SNR for Range, changing this db value can exclude or include more or less targets,

    Regards,

    Joe QUintal

     

  • Dear Joe-san

    We tried to make calibDcRangeSig enable, then it seems to be better.
    We will try to test more.

    Senoo.