• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 279 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.

Original question:

IWR1443BOOST: How to increase the number of point cloud data in 3D mode

IWR1443BOOST: Which demo should I use to detect a small target moving?

Lucas YANG
Prodigy 60 points
Part Number: IWR1443BOOST

Hello,

I wish to use IWR1443Boost to detect that whether a small target is moving, an ant for instance.

Could you please tell me which demo is recommended for achieving this?

Regards,

Lucas.

  • 0
    TI__Expert 7510 points

    Hello,

    Depending on the material of the object, the size of the object, and how far away the object is will determine if it is possible to get a return on the object. Below I am link a document which details radar performance for different sizes objects and discusses the results.

    mmWave Objects vs Range
    https://www.ti.com/lit/an/swra593a/swra593a.pdf

    Best regards,

    Connor Desmond

  • 0 in reply to Connor Desmond
    Prodigy 60 points

    Hi Connor,

    I want to detect some insects, ant for instance, with size 1cm, and the distance is about 50cm.

    (1) Is it possible to detect that whether ant is moving?

    By the way, I have also been trying to operate the OOB demo,

    (2) Is it possible to distinguish between 2 ants by watching the point cloud data? (For instance, one point refers to one ant movement trace, another point refers to another ant)

    I don't know is that only work for some materials with strong reflectivity, cuz I have did some tests and the result is not satisfied.

    Best regards,

    Lucas.

  • 0 in reply to Lucas YANG
    TI__Expert 7510 points

    Hello,

    The problem with this use case is that insects are small and not very reflective. This will mean the sensor will need to be very close to the ants. Even if you got the sensor close enough there is a limit to how close objects can get to the radar. Additionally, what you are battling here is the noise floor. The strength of the signal returned from an ant would be very small and would be buried by noise. The only way this might work is if the velocity resolution is small enough to be the speed of the insect. Also you will need to have static clutter removal on as well since you don't want to detect the surface the ant is moving along. In general though you will need to tune the radar to your use case and experiment and see what works and what doesn't.

    Best regards,

    Connor Desmond

  • 0 in reply to Connor Desmond
    Prodigy 60 points

    Hi Connor,

    Like you said, I turned on the static clutter removal. But while the radar facing to a static area where no object moving, the point cloud in the scatter plot are still changing and they change so fast. Are these due to the noise floor? Can we decrease it by operating the demo in somewhere with as few as reflective objects?

    Besides, I also wanna know how shall I change the velocity resolution value...

    Best regards,

    Lucas.

  • +1 in reply to Lucas YANG
    TI__Expert 7510 points

    Hello,

    I would try your experiments in an open area with as few objects in the FOV of the radar as possible. Please refer to the following resource about what determines the radial velocity. Also I am assuming that you are using the Demo Visualizer based on the image that you sent. The Demo Visualizer limits certain parameters of the device, but gives you the ability to visualize the results. If you don't want to be limited by the Demo Visualizer then you would need to provide a .cfg file as a piece of SW to parse/visualize the data from the device.

    I would suggest starting with larger objects and move to smaller and smaller and see what you can see. There is no guarantee that any of these suggestions will work, but you should definitely experiment to see where the limits are.

    Best regards,

    Connor Desmond