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.

Compiler/AWR2243: How to get point cloud, snr and noise data from AWR2243 cascaded RADAR

Part Number: AWR2243
Other Parts Discussed in Thread: IWR6843ISK

Tool/software: TI C/C++ Compiler

Hi TI Team,

We are using AWR2243 Cascaded RADAR in our application.
We would like to know how to get point cloud, SNR and Noise data (like demo applications (Area scanner, people counting) in industrial tool box) from the radar.

Is there any demo application available like single chip(IWR6843ISK) RADAR to get point cloud, snr and noise values?

Thanks
Sudharshan

  • Hi,

    No there is no such demo.

    With the Imaging kit one can capture raw data. With the Matlab post processing scripts provided in mmWave SDK 3.x you can create the point cloud.

    One can also write it's own post processing scripts

    thank you
    Cesar

  • HI Cesar,

    Could you please help to point out the matlab scripts in mmWave SDK 3.x(We are using mmwave_sdk_03_04_00_03), which will be used to get the pointcloud, snr and noise data from the AWR2243 cascaded device raw data?

    Is there any python scripts available for the same?If yes, could you please help to get the python scripts?

    Thanks

    Sudharshan

  • Hi,

    Sorry, my mistake, the scripts are not in mmWave SDK 3.x but in mmWave Studio 3.x

    In case you have not download it, here is the link

    MMWAVE-STUDIO-2G

    Please follow the instructions provided in the "mmwave_studio_cascade_user_guide.pdf"

    Thank you
    Cesar

  • Hi Cesar,

    We are using MMWAVE-STUDIO-2G.

    Could you please confirm the below path for scripts?

    C:\ti\mmwave_studio_03_00_00_07\mmWaveStudio\MatlabExamples\4chip_cascade_MIMO_example\


    Is this what you are referring?

    Could you please share the document for the usage of the script to get point cloud from 4chip_cascade raw data? Could you please share with example? It would be helpful.

    Is there any python scripts available for the same? It will help great.

    Thanks

    Sudharshan

  • Hi Sudharshan, 

    Please see the following e2e post for more documentation on the 4chip_cascade_MIMO_example Matlab example processing chain: https://e2e.ti.com/support/sensors/f/1023/p/911681/3426109#3426109. The signalPrcesssing_userguide_4chipCascade.pdf  is a draft of what will be released with the next mmWave Studio release soon. This walks through all of the Matlab examples. 

    We have no Python example code for the Cascade system at this time.

    Thank you,

    -Randy

  • Hi Randy,

    I didn't find the example for matlab application usage.
    Could you please share the steps How to get point cloud, snr and noise data from AWR2243 cascaded RADAR data with one example use case?
    This will help great.

    Thanks

    Sudharshan

  • Hi Sudharshan, 

    Please read through the \ti\mmwave_studio_03_00_00_07\docs\mmwave_studio_cascade_user_guide.pdf included in the mmWave Studio install for directions on how to setup the matlab processing examples. 

    The matlab example usage is described in section 8.3, but this section relies on data collection setup in previous sections. 

    Thank you,

    -Randy

  • Hi Randy,

    We are able to get the following data from the Matlab script

    xyz_all
    angles_all_all
    detection_results_all

    The xyz_all will represent point cloud data. What will detection_results_all and angles_all represents?
    How to get the Noise and SNR for these data?

    Thanks
    Sudharshan

  • Hi, 

    Detection results are the FFT points selected used for computing azimuth and elevation angles and ranges. 

    Angles are the estimated azimuth and elevation angles to these points.

    XYZ are the computed cartesian coordinates based on those estimated azimuth and elevation angles and ranges. 

    You would need to estimate a noise floor for the system first and then compare that against your detected target. I do not believe the example processing chain has this specific feature. Noise floor can be observed in a few different ways. This can be done in a clear scene (minimal clutter), or with a zero-slope chirp, or by observing higher-velocity 2D-FFT bins. 

    If that does not answer your question, can you please open up a new e2e post with this specific question?

    Thank you,

    -Randy