• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 102 subscribers
  • Views 379 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TIDEP-01017: Radar SDK usecases beamforming issue

Morris Huang
Intellectual 450 points
Part Number: TIDEP-01017

Hello,

I run the usecases "cascade_radar_beam_form", and the result is mismatched between GUI and real environment.

I think the reason is calibration of TX and RX.

Can i calibrate the TX and RX just like the usecsaes "cascade_radar_object_detect"?

According the source code of the usecases "cascade_radar_beam_form", the parameter can be  loaded from the text files.

The Radar SDK only provide the MIMO parameter text files.

Can TI provide the beamforming parameter text files?

Thanks.

Morris

  • 0
    TI__Genius 17200 points

    Hi Morris, 

    I am moving this to the ADAS processors forum to get you support. 

    Thank you,

    -Randy

  • 0
    Intellectual 450 points

    Hi,

    I try to calibrate the phase difference between the Rx at the beamforming mode.

    I put a corner reflector at 0 degree and on the static distance, and I measure the phase difference between each Rx for 10 frames.

    Then, i found the phase difference of each frame are different.

    According to my acknowledge, I think the phase difference between each Rx should be static if the corner reflector is on the static position.

    The result is shown below.

    Thanks.

    Morris

  • 0 in reply to Morris Huang
    TI__Genius 17200 points

    Hi Morris, 

    Phase calibration datasets can be effected by local operating temperature of the devices. Most consistent data will be gathered when the device remains at the same temperature. Can you repeat this experiment, but include the following steps: 

    1. Allow the system to heat-up to steady state ambient operating temperature with 30 seconds of dummy chirps before taking the actual calibration data?

    2. In the calibration dataset frames, discard the beginning of each frame until steady phase is observed 

    Can you graph what phase of the target reflector vs. chirp number looks like for your setup? It may help identify if these instabilities are being caused by temperature. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 450 points

    Hi Randy,

    I measure the Rx phase twice following these steps.

    step 1. I turn on the power of radar, and I let it do 60 seconds of dummy chirps.

    step 2. I record the data, and I turn off the power of radar.

    step 3. I go to step 1 after 60 second.

    Then, i found the phase difference of each frame are different.
    The result is shown below.

    Can you provide any advice?

    Thanks.

    Morris

  • 0 in reply to Morris Huang
    TI__Genius 17200 points

    Hi Morris, 

    Can you please show me the phase stability over each chirp. What does that trend look like inside of a given frame? I agree, the assumption is that the phase delta should be pretty stable between frames. Some drift is expected depending on how much the temperature varies over time. 

    Is the ambient temperature being controlled in any way? Is this an indoor or an outdoor measurement?

    Would it be possible for you to repeat this experiment in mmWave Studio using the TX-BF example project, comparing against the ProcessorsSDK code?

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 450 points

    Hi, Randy

    I try to measure the RXs phase for 1T16R, and the result is okay (<3 degree)(as shown as figure 1).

    I think the phase error is caused by TX beamforming in the demo.

    Can you have any advice?

    Thanks.

    Morris

    figure 1.