• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 1450 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: run vital sign lab on demo visualizer

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

Hi guys, 

How do I run the vital sign configuration file on the demo visualizer?

I tried to load the vital sign cfg file onto the visualizer, it gives an error as the last few lines of the vital sign cfg file are slightly different than the cfg files for the demo, plus vital sign uses 1 tx antenna while the demo file uses 2tx and 3tx antennas.

but is there any way to run the vital sign lab on the demo visualizer?

thank you guys

zhengguo sun

  • 0
    TI__Mastermind 28227 points

    Hello Zhengguo ,

    The vital signs lab uses a different signal processing chain than the out-of-box demo / demo visualizer. The OOB demo cannot extract the fine chest movements caused by the heart rate. As you may see in the documents published in this post, the format of the data output of these 2 demonstrations is very different, and the demo visualizer only works with the data format of the OOB demo.

    With regard to the configuration file, the vitalSignsCfg command that you noticed was rejected by the demo visualizer is meant to control the specific signal processing of the vital signs demo. The OOB demo cannot recognize this command and rejects it. At most, you could use with the demo visualizer the vital signs demo cfg file without the vitalSignsCfg command, but of course this would not display the vital signs.


    Regards,
    François.

  • 0 in reply to Francois Charlot
    Intellectual 425 points
    thank you so much for your explaination. Is there a way to simulate the vital sign lab? because however i change the configuration file, the results didn't show much difference.

    Also, is the default configuration the best values set for the vital sign lab? because it seems like the result doesn't seem accurate sometimes, and even with the same configuration, the results aren't very consistant, i.e. I would reset to the original configuration after I tested several different configurations as these configurations didn't do much on bettering the performance. But strangely, if I ran again using the original configuration, the result is very different than when I first ran with the same configuration.

    I'm trying to figure out a way to optimize the performance of the vital sign lab.

    Thank you for your help.

    best

    zhengguo sun
  • 0 in reply to ZHENGGUO SUN
    TI__Mastermind 28227 points
    Hello Zhengguo,

    It could be the inconsistencies you observe are due to the environment. As said in page 4 of the user guide of this demo, the radar should be pointed towards the person's chest and have an obstructed view to it. Also, the algorithm assumes the highest signal peak in the range profile corresponds to the person's chest; so if there is a large reflective object in the radar field of view that it could pick up instead of the chest, this would also be an issue. Some good tests of a proper setup are to verify the highest signal peak in the range profile plot correspond to the distance between the radar and the person's chest; then it is for the person to hold one's breath: after a few seconds, the GUI should display a breathing rate of 0, indicating the aim of the radar is correct.

    I have been using this demonstration for months and compared its heart rate estimation with that of a smartphone equipped with a heart rate sensor, and I found they very consistently match within 1-2 bpm. So I would say the radar configuration should be kept as is.


    Regards,
    François.
  • 0 in reply to Francois Charlot
    Intellectual 425 points
    thank you, Francois. In your response, you mentioned that ~ having the RADAR have "an obstructed view to the chest." It's sounded awesome, which is something I've been trying to achieve.

    As you may know, the field of view(FOV) for the awr1443 EVM is about 90 degree (obtained from arcsin(lambda/(2d)) where d=0.5lambda.) this is a very wide field of view. Altough most energy is focused around the boresignt, some interference from objects in the test environment is inevitible with such wide angle of view.

    So I'm trying to narrow down the FOV by activating the other TX antenna, which is the same way beamforming is done. But I coundn't do anything with it yet as I don't have many knowledge on programming. (Please share with me if you have a good idea on narrowing down the FOV of the RADAR, many thanks)

    Also, sometimes I would confuse "beamforming (multiple TX)" with "an improved angular resolution (multiple RX)" as there're some overlaps in their effect (I think both will INCREASE the RADAR's performance in accurately detecting and measuring its Target with reduced influence from Noise), what's the differnce between these two as far as their effect on the RADAR performance goes.

    Additionally, when a "waveform" is mentioned in a document, such as "... there are 256 or 512 points in the waveform, ..." do you mean the "IF signal waveform" or the "Chest Displacement Waveform"?

    Thank you so much,

    Zhengguo Sun
  • 0 in reply to ZHENGGUO SUN
    TI__Mastermind 28227 points

    Hello Zhengguo,

    I realize I made a typo and should have written unobstructed view; sorry for that. You could narrow the field of view by using a horn lens antenna. I cannot point you towards a specific manufacturer for fairness reasons, but you could find some on the web. This would be more straightforward than implementing additional signal processing techniques.

    Regarding your point about the 256 or 512 points in the waveform, I am not sure where you picked this sentence from as it does not seem to be from the vital signs lab. This may be related to the samples IF signal or to a plotted waveform like the range profile. You may want to close this thread and open a new one if this does not answer your point and this is not related to the vital signs demo.


    Thanks,
    François.

  • 0 in reply to Francois Charlot
    Intellectual 425 points
    Thank you so much for your help. It's been a very informing process.

    Zhengguo Sun