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.

IWR1443: Data array of radarCube

Part Number: IWR1443

Hi,

Why is the radarCube array in SDK 2.2.0.4 IWR1443 demo cmplx16ImRe_t?
From HWA UG swru526b.pdf Figure.19, I understand that the output array of HWA FFT is cmplx16ReIm_t.
(Figure.19, I = real, Q = imaginary)
In EDMA from HWA local memory to radarCube, it looks like there is no real / imaginary swap.

Please tell me why the radarCube array is cmplx16ImRe_t.

Best regards,
Hiroyuki Taguchi

  • Hi,

    Do you mean SDK 2.1.4?

    Thank you,

    Angie Mitchell

  • Hi Angie, 

    Sorry, the SDK is 2.1.0.4.

    Regards,

    HIroyuki

  • Hi Hiroyuki,

    The data here has the imaginary samples ahead of real samples. This is a Q ahead of I in raw data format. That will affect the data format in the range FFT and doppler FFT.

    Thank you,

    Angie

  • Hi Angie,

    Thank you for your reply.
    I understand as follows. Is it correct?

    1.

    The setting of adcbufCfg is
    adcbufCfg 0 1 0 1
    Since sampleSwap = 1, Q = LSB, I = MSB side,
    That is, the byte array of ADCbuf is as follows.

    ADCbuf address     data
    -----------------------------------
    4N                           Q[7:0]
    4N+1                       Q[15:8]
    4N+2                       I[7:0]
    4N+3                       I[15:8]

    2.

    By passing IQ data to HWA with this byte array,
    Complex FFT is executed on HWA with Q = imaginary, I = real.

    3.

    The byte array in HWA local memory (ACCEL_RAMx) of the FFT execution result by HWA is
    Same as ADCbuf in 1., as shown below.

    ACCEL_MEM address    data
    -----------------------------------------
    4N                                  Q[7:0]
    4N+1                              Q[15:8]
    4N+2                               I[7:0]
    4N+3                               I[15:8]

    Regards,

    Hiroyuki

  • Hi Hiroyuki,

    This is correct!

    Thanks,

    Angie