• State
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 36 views
  • Users 0 members are here
Support feedback
Options
Options
Related

IWRL6432W: Persistence noise point cloud on fixed distance (fixed intermediate frequency)

Wolfgang Jalma
Intellectual 303 points

Part Number: IWRL6432W
Other Parts Discussed in Thread: IWRL6432

Hi Champ,

 

I use a custom made IWRL6432 wcsp,and find when I use this chirp
chirpComnCfg 13 0 0 256 1 37 1
chirpTimingCfg 50 24 0 25 60
frameCfg 8 0 811 1 100 0

I found there are some persistence noise point clouds in the distance of 15 m (related to IF 2.5 MHz). I found when I change the chirp slope, those noise point clouds persist, change in distance, but still on the IF frequency of 2.5 MHz.

What are the causes, and how to deal with this problem?

Thank you

Sincerely,

 

Wolfgang

  • 0
    TI__Mastermind 29462 points

    Hello Wolfgang,

    Can you take screenshots and pictures of what you are seeing? I apologize but it is hard for me to understand what you are saying. Are you saying that when you change the chirp slope, you still see noise in the scene? Is there a reason you were expecting frequency slope to affect noise? I also do not understand where you are getting your beat frequency/IF as 2.5MHz.

    Are you using the mmWaveSensingEstimator for configuration guidance?

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Intellectual 303 points

    Hi Pedrhom,

    I am sorry that I have been not clear enough. We got  bellow setting from our customer, and run thorugh mmwave estimator, and found nothing wrong from it. The noise exist in form of point cloud as in figure 1, and when i look at range profile, I saw spike at 15 m, as in figure 2 Yes, I find when I change the slope, the noise is persist, but the noise moves closer when the slope is steeper, and further when the slope is flatter. When I input the slope and the distance into FMCW equation   , when S is Slope/Chirp rate, c is light speed and d is distance, I find IF frequency of the spike is always at 2.5 MHz. We think it either from EMI or from artefact from some signal processing. 


    So we ask experts here to share knowledge and/or experience to help us to understand why we get these noise and how  to solve this problem.


    Thank you very much.

    Sincerely,

    Wolfgang

    Config from customer noise at 15 m.cfg

    figure 1 Point cloude noise

    figure 2. Range Profile, there is power spike in 15 m, despite nothing in 15 m distance.

  • 0 in reply to Wolfgang Jalma
    TI__Mastermind 29462 points

    Hello Wolfgang,

    Thank you for clarifying, I am pretty sure I understand properly now. The "noise" peak moving in distance based on the config's range resolution is helpful information. It indicates to me that the issue is occurring consistently in the same range bins/ADC samples across chirp configurations, but due to the differences in maximum range and range resolution when adjust frequency/chirp slope, the peak is shown at a different distance but is consistently there.

    I notice the HPF filter being used is the 350kHz filter, can you try this configuration at 700 or 1400? Change the last field that is currently set to 1 in chirpComnCfg 13 0 0 256 1 37 *1* to a 2 or 3 for 700 and 1400 respectively.

    Also can you give a physical description of the environment these graphs were captured in? Was it in a low noise environment with a corner reflector? Was it in a typical room or outdoor space?

    Best Regards,

    Pedrhom