Hello everyone,

My name is Seungwon Lee, a graduate student in Korea. I am currently working on computing RCS (Radar Cross Section) using boundary E/H field data exported from CST T-Solver, and I have run into an issue that I have not been able to resolve despite extensive research through documentation, forums, and discussions with colleagues. I would greatly appreciate any insights or advice from those who may have encountered a similar situation.

Overview

We are post-processing CST T-Solver boundary E/H field data in MATLAB to compute RCS via near-to-far-field transformation, and we are observing a significant discrepancy between our MATLAB results and the farfield results computed directly by CST — specifically under the open boundary condition.

What Works: Open (add space) Boundary

As a validation step, we first tested our MATLAB post-processing pipeline using the Open (add space) boundary condition. In this case, our MATLAB-computed RCS results agree very well with the RCS results directly obtained from CST. This gives us reasonable confidence that our implementation is correct.

The Problem: Open Boundary (without added space)

When switching to the Open boundary condition (without added space), we observe the following:

• CST's RCS result covers only the angular range of −90° to 90°.
  This seems physically reasonable to us — since the scattering object is in direct contact with the open boundary and is effectively extended to infinity in that direction, no backscattered field is expected in the lower hemisphere.
• Our MATLAB result, computed by converting the boundary E/H fields into equivalent surface currents and integrating over all directions, produces data over the entire sphere (as shown by the magenta curve in the attached figures).
  This is also understandable, as the equivalence principle naturally yields radiation in all directions.
• However, in the overlapping range of −90° to 90°, the two results differ substantially, and we are unsure which one to trust.

Questions

I suspect that CST may use a different internal methodology for computing farfield results under the Open boundary condition — one that may not align with the standard near-to-far-field transformation. If anyone has experience with this, I would really appreciate your help with the following:

1. What farfield computation method does CST use internally under the Open boundary condition, and how does it differ from the conventional near-to-far-field transformation?
2. What is the physical reason for the large discrepancy in the −90° to 90° range?
3. Is there a recommended workflow or correction that should be applied when using boundary E/H fields from an Open boundary setup for external RCS post-processing?

Please refer to the attached figures for the simulation setup and results.

Thank you in advance to anyone who takes the time to read and respond. Any help or suggestions would be genuinely appreciated!

Best regards,
Seungwon Lee
Edu CST Studio SIMULIA