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.

AWR1642: CFAR threshold setting in mmw demo

Part Number: AWR1642

Hi champion,
   I am trying to adjust the CFAR threshold in AWR1642 mmw demo in mmwave_sdk_01_00_00_05. Could you help me to understand the euqation of threshold setting in the cfg file? I find in the user guide, the value threshold scale is explained as: Tclk = 256 * numVirtualAntennas * TdB / 6.  Could you help to check why we need divide by 6 in this equation?

Thanks,
Adam

  • Hi Adam, 

    After the 2D FFT is computed over a range gate, we need to run a detection algorithm over that range gate. The algorithms available all require the input to be in logarithmic form (not linear) i.e. in log2. For this we compute the log2Abs of the output vector, and then we run the LOG-CFAR-CA algorithm over it. The log2Abs has 8 fractional bits, hence the factor of 256. Since we non-coherently combine the output of all the virtual Rxs before running the CFAR, the additional factor of numVirtualAntennas is also present. Finally, to convert log2 to 20*log10, we need to divide by 6, as per the formula 

    TdB = 20*log10(|x|) = 20/(log2(|10|)) * log2(x)

    Regards,

    Anil

  • Hi Anil,
       Thanks for the reply. So if I understand it correctly, the CFAR threshold in dB wo usually mentioned is the dB value regarding the square magnitude of the received signal (|x|^2), in other words:

    TdB = 10*log10(|x|^2)
       Is it correct?

    Thanks,
    Adam

  • Hi Adam.

    That is correct.

    Regards,
    Anil