• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 31 subscribers
  • Views 627 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.

IWR6843ISK-ODS: Computes the range bias and rx phase compensation

Chen-Chih Liu
Intellectual 340 points
Part Number: IWR6843ISK-ODS

Hello,

I want to calibrate the range bias and Rx phase compensation in overhead_mount_occupancy with IWR6843ISK-ODS. I have found a source code "DPC_ObjDet_rangeBiasRxChPhaseMeasure()" which shown in "..\mmwave_sdk_03_05_00_04\packages\ti\datapath\dpc\objectdetection\objdethwa\src\objectdetection.c" to measure the range bias and phase compensation.

Can I leverage this source code into overhead_mount_occupancy? I know the DPC_ObjDet_rangeBiasRxChPhaseMeasure() uses an R-V heatmap to measure the range bias and phase compensation. I would like to know if I can use an R-A heatmap to find the range bias and phase compensation? I also know the memory allocation is different for radar cube data.

In addition to the above two differences, should I change others?

The purpose is that we want to calibrate each sensor in the future (such as on the production line) but we have our own bootloader (it means we cannot use OOB (out of box demo) to calibrate), so we want to leverage the DPC_ObjDet_rangeBiasRxChPhaseMeasure() into overhead_mount_occupancy or our own code to calibrate.

If we can leverage DPC_ObjDet_rangeBiasRxChPhaseMeasure()  but need to modify the input heatmap and numSymPerTxAnt (to fulfill DPIF_RADARCUBE_FORMAT_2), should we also need to modify the configuration file (vod_6843_ods_overhead_5zone.cfg)? I can directly set the parameters for distance (X) and window size (D) in the code, in other words, the CLI to set "measureRangeBiasAndRxChanPhase" does not need. 

Sincerely,

Jack

  • 0
    TI__Mastermind 30405 points

    Hi Jack,

    We will need to discuss with our algorithm experts whether Range-Azimuth heatmap can be used to compute the range bias and the phase difference between the antennas. Please note that due to holidays in the US, we will be able to get back to you early next year.

    Thanks for your patience.

    Regards

    -Nitin 

  • 0
    TI__Mastermind 40025 points

    Hi, Jack:

    The range bias and phase calibration needs to compute the zero Doppler first.  But the overhead people counting demo remove the static clutter (zero Doppler) information.  Therefore, we can not use the current overhead people counting demo for range bias and phase calibration. 

    In the current OOB demo, function "DPC_ObjDet_rangeBiasRxChPhaseMeasure()" is called after Doppler FFT.  Since Doppler FFT is not called after range FFT in the overhead people counting demo.    I would suggest to create another function after range FFT operation, and before clutter removal to calculate the zero Doppler signal (instead of the complete the Doppler FFT) and called the modified DPC_ObjDet_rangeBiasRxChPhaseMeasure() function after that.  

    Let me know if it makes sense to you.

    Best,

    Zigang

       

  • 0 in reply to zigang Yang
    Intellectual 340 points

    Hi Zigang,

    The phase could be calibrated and the results look well.

    Thanks for your suggestion.

    Best,

    Jack