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.
Hello,
I have some questions about clutter removal function of AWR1642.
I understand calculation procedure of clutter removal.
(The mean value of the input samples to the 2D-FFT is subtracted from the samples.)
Although removal static clutter is enaled, target that have zero doppler index is detected sometimes.
Please see the picture below.
In our office, when AWR1642 is not moving (AWR1642 is put on my desk), there is no detected object, because there is no moving object around AWR1642.
But once I move AWR1642 , since there are relatively moving objects, some detected objects appear, and
in range profile for zero doppler plot, there is one detected object.
(I think it is spectral leakage of FFT.)
Do you have any ideas why target at zero doppler is detected in spite of enabled clutter removal?
SDK: SDK2.1
C:\ti\mmwave_sdk_02_01_00_04\packages\ti\demo\xwr16xx\mmw\xwr16xx_mmw_demo.bin
mmWave demo visualizer: 3.1.0
Hi,
I will need a few days to check with our systems team for theoretical explanation
thank you
Cesar
Hi,
I have discussed your observations with the systems team and we reached the following conclusions:
The Clutter Removal is not designed to be used when the sensor is moving.
Clutter Removal is designed to be used in an environment where
When sensor is moving the zero-Doppler bin no longer represents the Static clutter in the scene. Because of the relative movement of the sensor with respect to the other objects in the scene, these other objects would appear at different Doppler bins. Enabling Clutter Removal does not make sense
thank you
Cesar
Hi,
Thank your for your answer.
I understood when sensor is moving, there are only moving objects around the sensor and in that environment clutter removal does not make sense.
But in that environment, I think since there is no static object originally, object that has zero doppler index is not detected, but in the above figure, zero doppler object is detected. Please see 'range profile for zero doppler' plot.
Do you have any ideas why it occurs?
You are correct, clutter removal does not make sense in this environment.
In addition, THERE IS NO MORE ZERO DOPPLER. Since the sensor is moving what used to be zero doppler when the sensor was not moving is now in a doppler bin and detection will happen
Does this answer your question
thank you
Cesar
Hi,
Thank you for your answer.
I think I am not fully understanding.
Please see 'range profile for zero doppler' chart that I posted first.
If clutter removal is enabled, I think there is no 'orange dots' in the chart, because orange dots mean detected object and there is no static object when sensor is moving. But the figure I posted, there is a orange dot (detected object) in the chart.
Do you have any ideas why it occurs?
Hi,
My understanding of the experiment is as follows:
1) the board is moving
- with moving board clutter removal does not matter.
2) there is one object detected.
This is expected behavior. Why don't you expect this?
Thank you
Cesar
Hi,
Thank you for your answer.
Regarding 2), I can’t understand why objects that have zero doppler are detected. Though clutter removal doesn’t make sense, there is no zero doppler object.
For example, is there possibility that object detection on zero doppler index is caused by spectrum leakage from non zero doppler object?
Hi,
I think I understand now your question.
Let me discuss with systems team and get back to you
Cesar
Hi,
I disabled clutter removal(all objects can be detected), and performed the experiment with moving EVM.
Then some objects that have zero doppler index were detected.
(There are some orange dots on 'Range Profile for zero doppler' chart)
Hi,
I have discussed these observations with the systems team.
We think that these detections are caused by leaking from strong reflections
thank you
Cesar