• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 503 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.

IWR6843ISK: How to improve the resolution of IWR6843ISK

Xuran Sun
Prodigy 40 points
Part Number: IWR6843ISK

Hello,

My name is Xuran. I was trying to achieve the demo of fall and stance detection. But I noticed that the resolution is low and there were very limited points showed in the GUI. It is really hard to get the outline and the shape of a person in the room. The possible reason may be the device. The demo uses IWR6843ISK-ODS, but I used IWR6843ISK. So, I want to ask is there any possible solutions to improve the performance of this experiment? How about changing some parameters of the configuration file? If it is yes, which parameters can be changed to get a better performance? Could you give me some tips? Thanks in advance! 

sensorStop
flushCfg
dfeDataOutputMode 1
channelCfg 15 7 0
adcCfg 2 1
adcbufCfg -1 0 0 1 1
profileCfg 0 60.6 500 7 40 0 0  50 1 128 5300 0 0 30
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 2
chirpCfg 2 2 0 0 0 0 0 4
frameCfg 0 2 16 0 100 1 0
lowPower 0 0
guiMonitor -1 1 0 0 0 0 0
cfarCfg -1 0 0 8 4 4 0 5120
cfarCfg -1 1 0 4 2 3 0 5120
peakGrouping -1 1 1 1 1 255
multiObjBeamForming -1 1 0.5
clutterRemoval -1 1
calibDcRangeSig -1 0 -5 8 256
extendedMaxVelocity -1 0
compRangeBiasAndRxChanPhase 0.0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0
measureRangeBiasAndRxChanPhase 0 1.5 0.2
sensorStart

Regards,

Xuran Sun

  • 0
    TI__Mastermind 28055 points

    Hi Xuran,

    First, there is an important difference between ISK and ODS configurations (due to software)

    ODS chirp configs:

    chirpCfg 0 0 0 0 0 0 0 1
    chirpCfg 1 1 0 0 0 0 0 2
    chirpCfg 2 2 0 0 0 0 0 4

    ISK chirp configs:

    chirpCfg 0 0 0 0 0 0 0 1
    chirpCfg 1 1 0 0 0 0 0 4
    chirpCfg 2 2 0 0 0 0 0 2

    This sets the order that the transmitters chirp in to be different - this will help a lot with the issue you are seeing.  You can also use the online mmWave visualizer to download a chirp config with high range resolution and high doppler resolution. (You can use the sliders on the GUI to change these, then download the configuration to the PC.)

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Prodigy 40 points

    Thanks Justin

    I have tried to modify the sets of the transmitters chrip. But the case was not really much better than before. And could you please attach a link where I can get the high range resolution and high doppler resolution configuration? 

    Regards,

    Xuran Sun

  • 0 in reply to Xuran Sun
    TI__Mastermind 28055 points

    Hi Xunan,

    If there is an ISK pre-built binary, ensure you are using that one. You can use the OOB demo here, that is what was used in the Fall Detection experiment. You can get chirp configurations with better resolution here: https://dev.ti.com/gallery/view/mmwave/mmWave_Demo_Visualizer/ver/3.3.0/

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Prodigy 40 points

    Thanks for your reply, Justin 

    The issue now is that the device is not able to detect the object which is 2 meters away from the sensor or only a few points near the ground. I think the detection range should be 14.3 meters according to the parameters in the configuration file.

    Additionally, when I sitting on the chair, some high points which should only appear under the standing situation also appear, I am not sure whether is the issues of the device or configuration. Could you give me some hints which could resolve the problem? 

    Regards,

    Xuran Sun 

  • 0 in reply to Xuran Sun
    TI__Mastermind 28055 points

    Hi Xuran,

    The ISK has a narrower Field of view than the ODS. So you should make sure anything you are trying to detect falls inside the FOV of the device. Secondly, the ISK has less angular accuracy in the elevation direction than the ODS - this could lead to points appearing above your head.

    Additionally, please ensure that peakGrouping is disabled in your chirp configuration. This is a cli command, see the SDK User's guide for information on its use.

    Do you have any pictures or video showing the issues with detection here? It would be helpful to see where the object is in relation to the sensor.

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Prodigy 40 points

    Hi Justin

    I don't know why the video can not be uploaded. So I have uploaded several screenshots. The first screenshot shows the time point when the person walks away from the sensor. The location is appropriate satisfactory but only blue points presented, which means the higher points (head) is not showed or detected. However, you can see in the second screenshot, it is normal, the height basically matches the height of the person. The issue is that the farther the person away from the sensor, the fewer points presented and the lower height of the points.

    Regards,

    Xuran Sun

  • 0 in reply to Xuran Sun
    TI__Mastermind 28055 points

    Hi Xunan,

    "The issue is that the farther the person away from the sensor, the fewer points presented and the lower height of the points." - How is the EVM mounted? Is it tilted down? How high is the device off the floor?

    From your description that I quoted above, it sounds like at different ranges different parts of your body are in the sensors Field of View. The ISK has a 40 degree elevation FOV. If you are close to the sensor, only part of your body will fall into that 40 degree area - could this be why you only see the higher points associated with your head when you get close? When you are farther away, you are only seeing your feet - if the sensor is pointed down, this is where the FOV would fall at range.

    Try setting the device at roughly 1.25 meters (chest/abdomen height). Make sure the sensor is not tilted down or up - boresight should be parallel with the ground.  This should give a good detection of your body, especially when walking.

    Regards,

    Justin