• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 808 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.

MMWCAS-RF-EVM: Question on calibration process of MMWCAS-RF-EVM

Hang
Intellectual 765 points
Part Number: MMWCAS-RF-EVM

Hi, 

I have two questions related to the calibration of cascaded radar (MMWCAS-RF-EVM) on page 15 of Imaging Radar Using Cascaded mmWave Sensor Reference Design. 

(1) The I_interp or calibrationInterp is given as 5 in file module_param_antennaCalib.m under C:\ti\mmwave_studio_03_00_00_14\mmWaveStudio\MatlabExamples\4chip_cascade_MIMO_example\main\cascade\paramGen. 

The interpolation factor (calibrationInterp) is used for range FFT for frequency calibration.

Why it is given as 5? For different chirp configurations, how to chose the value of calibrationInterp?

(2) On page 15 of  Imaging Radar Using Cascaded mmWave Sensor Reference Design, exp(-jF) should multiplied with the raw ADC data for frequency calibration. 

However, in file datapath.m under C:\ti\mmwave_studio_03_00_00_14\mmWaveStudio\MatlabExamples\4chip_cascade_MIMO_example\modules\calibration\@calibrationCascade, line 108 "correction_vec = (exp(1i*((0:numSamplePerChirp-1)'*freq_calib))');", the negative sign of exp(-jF) is gone. This is inconsistent with the description. 

Where is the negative sign of exp(-jF) in the calibration process or which one is correct?

Thank you very much for the help!

Best,

Hang

  • +1
    TI__Genius 13805 points

    Hi,

     Please see the response below.

    1) 5 is a just an example number used to increase the FFT bin size to increase the frequency calibration accuracy

     2) correction_vec = (exp(1i*((0:numSamplePerChirp-1)'*freq_calib))');

    The transpose sign highlighted in bold and larger size for complex vector is taking the conjugate, so it is adding the minus sign already.

    Or you can write in the following way.

     correction_vec = conj(exp(1i*((0:numSamplePerChirp-1)'*freq_calib)).');

    Regards

  • 0 in reply to Abhed Misra
    Intellectual 765 points

    Hi Abhed,

    Thanks a lot! This solves my questions. I have another question on how to choose calibrationInterp, as you said it was used to increase the FFT bin size. Are there suggestion values for different FFT bin size, like, 32, 64, 128, 256, 512, 1024?

    Thanks!

    Best,

    Hang

  • 0 in reply to Hang
    TI__Genius 13805 points

    Hi,

          The bin size is linked to your modulation scheme and target resolution(at target range) along with Scan zone(angle). You can decide to tune the bin size accordingly.

    Regards

  • 0 in reply to Abhed Misra
    Intellectual 765 points

    Hi Abhed,

    Thank you very much!

    Best,

    Hang