Using Sionna for synthetic aperture radar simulation: variance in path propagation delays #463
Replies: 2 comments
-
Hi, By reducing the distance between the radar and the reflector (I reduced |
Beta Was this translation helpful? Give feedback.
-
Converting this to a discussion as it appears there is no issue with Sionna. |
Beta Was this translation helpful? Give feedback.
-
Hello,
first and foremost, thanks to Sionna's authors for this amazing tool. I would like to use it for synthetic aperture radar (SAR) simulations. I am aware that radar simulations is not exactly within Sionna planned scope, but I would like to know if there is hope for my experiment or not.
In synthetic aperture radar, for a single point reflector, the received signal exhibits a quadratic phase in the azimuth dimension as the platform moves through the synthetic aperture, because the range to the target first decreases (until the point of closest approach) then increases. As a first sanity check, I would like to check wether or not this is true with the raytracing component of Sionna. I wrote the following code, which is as close to the scattering tutorial as possible:
After running this I get the following plot:
![sionna_sanity_check](https://private-user-images.githubusercontent.com/64254772/334177184-f625f035-7854-45b8-a380-73fc3b9ca2a3.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.ou_4vcOmcj1wg4ToDNpFJcehvuqTmu4hflXtBBDgCrU)
which shows a significant difference between the theoretical quadratic range and the average length of rays to the reflector. While those differences are small with respect to the overall length of the ray (less than 10cm over a 500m long ray), the carrier frequency term in the propagation phase of the SAR signal model leads to large differences in phase between both range models.
Is this an intrinsic issue with the inner workings of sionna and mitsuba which is not easily fixable or is there something that I can do to reduce this range "error" ?
Thanks in advance to anybody who can provide insight into this issue. Please ask if anything is unclear.
Beta Was this translation helpful? Give feedback.
All reactions