• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 768 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: Antenna BR measurement Improvement

ZHENGGUO SUN
Intellectual 425 points
Part Number: AWR1443BOOST
Other Parts Discussed in Thread: AWR1443

Hello everyone,

I've been using the AWR1443 EVM for our vital sign project for some time now. During the experiment, we tried to simulate the real-world workflow and mounted the antenna on the side rail of the patient bed, which means the subject isn't positioned orthogonally to the transmitted signal.  But given there's a 60 degree of Azimuth field of view for the board we are working with, the antenna still detects the subject's chest motion; however, the breathing-rate resulted isn't consistently accurate; this data constantly fluctuating through a wide range, such as single-digit to above 20.

I'm wondering if there's any way to improve the accuracy and consistency of the Breathing rate measurement for our experimental setup.  This setup is depicted in the figure below.

Thank you so much,

  • 0
    TI__Guru 50240 points
    Hello again Zhengguo,

    That is a very clever idea to turn the EVM on its side to enhance elevation field of view!

    Can you confirm you doing the same demo as the Vital Signs Lab from the mmWave Training? If not, how is your version different? Can you share your chirp configuration if it is different?

    If you can answer these questions I can help you optimize your configuration of the sensor for your particular scene; moreover, a picture of the scene might also be helpful if possible (no problem if it isn't).


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

    thank you for your response.

    Actually, it is the same vital sign lab. But the .cfg file may look a bit different. which I copied below.

    • sensorStop
    • flushCfg
    • dfeDataOutputMode 1
    • % channelCfg 15 5 0
    • channelCfg 15 2 0
    • adcCfg 2 1
    • adcbufCfg 0 1 0 1
    • profileCfg 0 77 7 6 57 0 0 70 1 100 2000 0 0 40
    • chirpCfg 0 0 0 0 0 0 0 1
    • %chirpCfg 1 1 0 0 0 0 0 4
    • frameCfg 0 1 2 0 50 1 0
    • guiMonitor 0 0 0 1
    • vitalSignsCfg 0.3  1.2  256  512  0.1  0.1  0.05  10000  10000
    • sensorStart

    the lines I commented out using "%" symbol is the tentative .cfg to enable the two TX antennas that are seperated by 2lambda. But with 2 TX .cfg file the GUI does not work (froze).

    also, when I changed the guiMonitor line to 0  0  0  0, the GUI still works. so I'm wondering if disabling all monitors by using 4 zeros in the guiMonitor line would improve the GUI performance as I feel the number of monitors enabled directly affects the data rate or the internal computation of the algorithm or the software performance (Honestly, I can't explain what I'm trying to say here as I can only recall that i read this relationship off of one of the Whitepaper or some paper that is in pdf format)

    And I'm sorry that I couldn't provide the picture for the particular scene depicted in the picture in my original post as I didn't have one myself. But I hope the picture in the original post would help.

    Thank you so much,

    Zhengguo Sun

  • 0 in reply to ZHENGGUO SUN
    TI__Guru 50240 points
    Hello again Zhengguo,

    The extra antenna is not needed for this lab since we are looking at single object (person) at a predictable distance (.3 to 1.2 meters) and simply seeing how much chest displacement they are displaying. Futhermore since the extra enabled antenna is not used in this lab so when you are enabling it you are likely interfering with multiple levels of both processing and detection. I would not recommend configuring additional antenna for this reason.

    The guiMonitor line is likely from a previous build in a completely different gui so no editing is going to be necessary here.

    If you are looking to improve gui performance, keep in mind that all of our data is coming through UART. On a standard PC the UART is the one of the least prioritized peripherals running so it would be wise to close out of as many other programs as possible (including CCS if possible).


    Cheers,
    Akash