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: Real-time experiments with Python

Part Number: AWR1642

Hi,

     We have AWR1642+DCA1000, and we are interested in real-time data processing and visualization in Python, and we would like to do hardware in the loop experiments

     1. We can use mmWave demo visualizer, but we are able to get very low (<3 fps) range-velocity heat maps (Serial port has low data bandwidth)

     2. We can also use mmWave Studio GUI to record and post-process recorded data, but this is not suitable for hardware in the loop tests

     3. For rapid prototyping of high-level ideas, coding in C seems to be time consuming

We have developed a solution, but we would like to know whether there is a more practical solution (i.e. are we re-inventing the wheel or not).

I will appreciate getting some feedback about the following experiments where we process radar frames at 25 fps, do 2D FFT, do some openCV stuff, and do visualization (All real-time). Note that frame = 128 chirps, chirp = 256 samples, and we have 4 receive antennas. From a programming perspective, the mmWave radar is used as if we are using a live video capture device.

https://www.youtube.com/watch?v=2kEqh4AuMGo

https://www.youtube.com/watch?v=MDlA2KCt7BA

https://www.youtube.com/watch?v=jNuBgl6KMCs

  • Hi Onur,

    We don't have any official solution to do the real-time processing using DCA1000, although in this forum many users have tried doing this experiment. I think you also did live processing using DCA1000 (based on video).

    By default with DCA1000 you receive raw ADC data which will be in big size and utilize much ethernet bandwidth. To reduce this, you can send only 1-D or 2-D FFT result (or any other intermediate result) instead of raw ADC data which you can read on PC using Python or any other scheme to do further high level processing. This way you can easily achieve higher fps (faster) without loosing much data.

    Regards,

    Jitendra