• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 1200 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

AWR1443BOOST: visualizer, number of object detected, loops of chirp

ZHENGGUO SUN
Intellectual 425 points
Part Number: AWR1443BOOST

Hi,

In my experiment with demo visualizer, the 2 tx antenna cfg file detected about 2 times more objects than the 3 tx antenna cfg file with the same profilecfg and framecfg setting, i.e. for both 2tx and 3tx configuration files, I changed the profilecfg line and the framecfg line so that they both are configured as follows:

profileCfg 0 77 7 6 57 0 0 70 1 100 2000 0 0 30

frameCfg 0 1 20 0 50 1 0

and the result is that with 2 tx antenna, the number of the detected object is about 100 while the result for the 3 tx antenna is 51. this confused me a little, shouldn't it be the other way around? 

Also, the number of object detected increased with the number of loops or chirps per frame (i.e. the 3rd number in the frameCfg line). could anyone explain to me why is such relationship and is it better to have more object detected per frame? does it mean a better range resolution? However, the result for the range resolution number didn't change at all throughout my experiments with the demo visualizer. 

thank you very much.

zhengguo sun

  • 0
    TI__Guru 50240 points
    Zhengguo,

    Apologies for the delay. More Tx or Rx antennae are going to contribute to an improvement in angular resolution. This means more angle bins which is going to result in more objects being detected in the same range bin.

    It is very odd you see less objects with more Tx antenna. Can you take a picture of your setup? Perhaps try changing the setting completely to see if this is still the case?


    Cheers,
    Akash
  • 0 in reply to Akash Gondalia
    Intellectual 425 points

    thank you for your answer. Today I experimented again, and it seems like the RX gain parameter (the last number on the profileCfg line) is directly related to both the profile curve and the noise curve on the range profile plot, that is, when I increase the RX gain parameter both curves shifted up, and the opposite happened when I reduced the RX gain parameter.

    Another observation is that the number of loops per frame (the 3rd number on the frameCfg line) is inversely related to the noise level. In my experiment, as I increased the number of loops per frame, the noise curve on the range profile plot shifted down, and a further increased loop counts further reduced the noise level.

    figure 1. Blue=rangeProfileCurve. Green=NoiseCurve. (Cfg info. RXgain=26. Loops/frame=10)

    Figure 2. cfg info. RXgain=26. Loops/frame=38

    Figure 3. Cfg info. RXgain=46. Loops/frame=10

    Figure 4. Cfg info. RXgain=46. Loops/frame=38

    So I am wondering

    1. if the more number of chirps (or loops) per frame I have the better antenna performance I can achieve (low noise).  Also,

    2. Should I keep the RX gain parameter as small as possible? because a larger RX gain parameter would increase both the range profile curve and the noise curve, if the range profile reflects the signal strength of the antenna, then the increase(up-shift of curves) in both signal and noise will, in fact, result in a smaller signal to noise ratio.

    I don't know if my observations are any misleading.  Please help me improve.

    Also, in my previous post, I mentioned the relationship between the number of loops per frame and the detected object counts. they demonstrated a directly proportional relationship to me, but I'm not certain if this relationship is true, and if true why is that.Moreover, I don't understand how come the number of objects detected increased drastically if there aren't as many objects inside the room I experimented with the EVM.  

    thank you so much

    zhengguo sun

    (P.S. about the object counts with 2TX and 3TX experiment, I will run the demo again tomorrow, and I'll capture the result then send it to you. thank you for your help)

  • 0 in reply to ZHENGGUO SUN
    TI__Guru 50240 points

    Hello again Zhengguo,

    1. if the more number of chirps (or loops) per frame I have the better antenna performance I can achieve (low noise).

    This is incorrect, if you want a lower noise floor you need to increase chirp duration. More chirps per frame will inherently lower your chirp duration.


    2. Should I keep the RX gain parameter as small as possible? because a larger RX gain parameter would increase both the range profile curve and the noise curve, if the range profile reflects the signal strength of the antenna, then the increase(up-shift of curves) in both signal and noise will, in fact, result in a smaller signal to noise ratio

    This is correct, Noise Floor increases proportionally with Rx Gain. To optimize this you are going to want to reduce your Rx Gain to a number that is acceptable for your application.

    Let us know if you have any more questions at this time!

     

     

    Cheers,

    Akash