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.

AWR1243BOOST: AWR1243BOOST, Compensate Rx Channel Gain/Offset

Part Number: AWR1243BOOST
Other Parts Discussed in Thread: AWR1243

Dear experts,

I'm using AWR1243BOOST to capture raw data, and doing post processing in MATLAB.

It seems there is a variation of amplitude and phase in received signal of each Rx channel, which needs to be compensated.

I could find the document about XWR16XX in mmWave SDK documents,

and in the case of  XWR16XX, I understood that there is compensation method as Rx Channel Gain/Offset.

Now, how can I measure and compensate it in my own post processing with AWR1243?

If you have any suggestions or technical documents, would you please let me know?

Regards,

Masa 



  • mmWave device provides best possible (compenseted) ADC data. But still you can refer mMWave SDK features which compensate further. Refer these two CLI command implementation in mmw demo as reference: measureRangeBiasAndRxChanPhase, compRangeBiasAndRxChanPhase

    There is basic Matlab script to read the ADC data from captured file (section 9)
    www.ti.com/.../swra581b.pdf

    Regards,
    JG
  • Thanks JG,
    The Matlab script of the document is very helpful. Thanks to it, I could do post processing very smoothly.

    Is the command of "compRangeBiasAndRxChanPhase" effects raw ADC data?
    I mean I understood that the command effects on-chip processing, and it is not useful for raw ADC capture.

    Is my understanding not correct?

    Regards,

    Masa

  • Hi Masa,
    Your understanding is correct, this command has an effect on on-chip processing, not on raw ADC data.

    Regards,
    Jitendra
  • Hi, Jitendra
    Thank you for your response.

    So again, my first question is
    "How can I measure and compensate Rx Channel Gain/Offset in my own post processing with raw ADC data of AWR1243?".

    If you have any technical suggestions, it would be very helpful.

    Regards,
    Masa
  • Hi Masa,

    1. Collect data as usual with a corner reflector placed directly opposite the radar (at Azimuth angle '0').

    2. In your matlab script, after the 2D FFT, identify the range-bin and the doppler-bin corresponding to the corner reflector. Extract the samples of that particular bin for all antennas. Let this vector  be 'v'.

    3. Normalize the vector v.

    v = v/abs(v);

    4. Ideally, all antennas should have roughly the same phase (since the target is at 0' Azimuth). If that isn't the case, we should correct all the phases, so that they are equal (or zero). Therefore the correction parameter is simply conj(v).

    Regards

    Anil