search

ekurtgl / FMCW-MIMO-Radar-Simulation

192 stars 37 forks source link

Cartesian to Polar Distortion towards the FOV edges #4

Closed DenizEDRS closed 3 weeks ago

DenizEDRS commented 1 month ago

Hi, im currently working on a similar program. The issue i am encountering is that artificially created Range-Angle Heatmaps are not fully accurate. In your animation at the very bottom you seem to have a distortion that occurs towards your angular edges, do you know how those are describable to be reproduced?

ekurtgl commented 1 month ago

Hi,

It is an expected phenomenon as the target approaches closer to the radar, its body parts will allocate more bins in the angular spectrum. You can think of a small rubik cube example. If the cubical object is at 0.1 m away from the radar, its return signals can allocate the angular intervals of -45 to 45 degrees, but when it is 5 m away, it might be detected only at -5 to 5 degrees. We observe a similar phenomenon here. So, it is not a distortion, but an expected outcome. I hope this helps.

DenizEDRS commented 3 weeks ago

thank you for the quick answer, i am now able to replicate radar heatmap generations, the issue was due to range clipping in my own code that wasnt fully adjusted.