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.

IWR1642: Regarding the frameCfg and subFrameCfg parameters

Part Number: IWR1642

Hello TI employees.

I have some questions on the parameters included frameCfg and subFrameCfg,

1. In frameCfg, the third parameter can be used in setting number of loops. In the mmwave_sdk_user_guide, this parameter can be set range from 1 to 255.

But the value have to be power of 2, so settable maximum value is merely 128. I checked the IWR1642Boost worked properly when I set the vale as 128. Am I right? 

I want to confirm whether the maximum configurable value of "number of loops" parameter is 128 or not.

2. To my knowledge, the value of number of loops parameters is related to the number of doppler-FFT.

We can get finer velocity resolution if we set the higher value. The question is right here.

If the answer of the Q1 is yes, we cannot  get any more finer velocity resolution.

Is there other way to enhance the velocity resolution simply by modifying some parameters in the current mmWave Demo?

3. In subFrameCfg, there is also numLoops parameter. The description table of the document says the value of this parameter is not constrained.

Can this parameter be used instead of the "number of loops parameter" in the frameCfg.

Here is the end of my questions.

Thank you.

  • Hi,

    Q 1 and 3 will need more investigation.

    For Q2, the velocity resolution depends on doppler bin which is equal to numChirpsPerFrame / numTxAntennas. 

    numChirpsPerFrame  is (frameChirpEndIdx -frameChirpStartIdx + 1) *numLoops;

    If more chirp profiles are defined, velocity resolution can be improved.

    I'll get Q1 and 3 answers as soon as possible.

    Regards,

    Michelle

  • The reason that numloops or numDopplerBins need to be kept power of 2 is that there is no zero padding for 2D FFT. As discussed previously, numchirps per frame can be increased to increase numDopplerBins as long as it's kept to be power of 2.

    As to your Q3, subframe numloops has the same requirements as frameCfg.

    Regards,
    Michelle
  • Thank You Michelle Liu
    You helped me a lot.

    Have a good day.