Odd measurements on python example (BGT60TR13C)

[JoKa]

Hello, 

I have been trying to get accurate readings using the Python examples for the BGT60TR13C card. When placing a corner reflector at a distance of 1m, the static distance example works well and measures 1m while the FFT distance program shows a peak at 0.8m. Is the settings off? What is wrong? 

I would like to measure the range-angle relation as well. I have my corner reflector in an otherwise isolated room, and still the measurements are all over the place. Is it the program that is not working? I have isolated the cable and everything, there should be no other clear reflection in the room. 

I'll attach a picture for you to consider.

Regards

[Yashraj_P]

Hi @JoKa ,

1. "distance_fft.py" and "static_distance.py" use same algorithm to obtain distance, so I think there should be not much difference in their output. But in the example code "distance_fft.py" measures distance for all 3 receivers and outputs the same. While "static_distance.py" has rx_mask set as 1 so, only 1 receiver output is seen. Is it 0.8 for all three antennas in "distance_fft.py".
2. As for "range_angle_map.py" example, it works well for us, in the office environment. We are working on to find why it is behaving erratically on your end.

Additionally, what would be the RCS of the corner reflector you are using?

Thanks and regards,
Yashraj

[Yashraj_P]

Hi @JoKa ,

We replicated your experiment with Corner reflector, but did not see any errors, following is the output we got. 

We would like to you to use the sensor in landscape mode as following, with the range_angle_map.py example.
Could you please share the script you used and the console output again with a picture of your setup if the error is still not resolved.

Thanks and regards,
Yashraj

[JoKa]

Hello, 

Thank you for your response. For FFT all three antennas show a different value on the graph than even in its own program, I'll attach a picture where you can see the antenna readings as well as the diagrams. I'm using the example code, I've only changed the max range to fit the setup. 

Thank you for trying my setup, its a 4 cm corner reflector so the RCS should be well enough. I've changed the antenna to face the way you suggest, but it still has a lot of noise. However when I wiggle the cord while measuring it seems to do better? In my setup the reflector is at a distance of 1m, 20 degrees from normal. The program reads 60 cm distance and 10 degree angle. I tried for different distances and it was always roughly 30-40 cm off in distance and 10 degrees of in angle. I'll attach pictures. 

[Yashraj_P]

Hi @JoKa ,

Seeing your setup I suspect the boundaries of the mounting structure (which holds the BGT60TR13C radar board), is reflecting and interfering with the signal. And this might explain the bad output.
I would like you to perform the experiment by placing the radar on a tripod (or something else), with nothing blocking its view. 
Please share the picture of new setup, if the suggestion does not work. And also the SDK version you are using for the example code.

Thanks and regards,
Yashraj

[JoKa]

Hello,

Thank you for your response. What has me confused is that the static distance and the FFT SDK programs show different readings, regardless of the setup they should be the same?

I feel like same same goes for the range-angle map? The static distance will show 1m (the correct distance), FFT will say 80 cm and the range-angle will show roughly 60 cm. 

I downloaded the RDK-TOOLS-SW 2022-08-24 and use the SDK examples for Python. I have also copied the radar_sdk.dll and ifxRadarSDK.py to the directory of the example. 

Regards, 

[Yashraj_P · Accepted solution]

Hi @JoKa ,

Can you confirm that the response you mentioned is seen by other samples of BGT60TR13C?
We checked the application with boards present with us and found that the applications are working as expected with proper setup.

Also the sample examples, static_distance.py, distance_fft.py and range_angle_map.py use different combinations of receivers. This can result in different output based on setup. Please consider the measures to be taken as mentioned in my previous reply.

Moreover, the applications we built are just example codes and serve as starting point for customer to retrieve data from Radar and process it. You could always refine it make it more robust.

Thanks and regards,
Yashraj

[Yashraj_P]

Hi @JoKa ,

We will be locking this thread  due to inactivity. If you have further queries, please create a new thread.

Thanks and regards,
Yashraj