Hi Mariangela,

Ray tracing is a good method for determining coupling and other parameters so long as as diffraction effects are unimportant. Generally this means that feature sizes are large with respect to wavelength and that self-diffraction in a coherent beam is also unimportant.

For such systems there are a number of ways of evaluating beam characteristics. A Detector Polar can be used, and not just with sources. It should be positioned with its origin at the apparent source, and for an extended source its radial size should be made much greater than the size of the source. A Detector Rectangle can also be used. It can be set to display in either position or angle space. In optimization, the NSDD operand can also measure in angle space as well as position space. For example, it can be set to report the centroid locations as well as the RMS sizes in angle space.

Having said this, I am concerned about your design being quasi-optical sub-THz, and using a horn antenna. In my meager experience, such implementations often do not meet the requirement that feature sizes are much greater than a wavelength. OpticStudio does have capabilities for working with such systems using physical optics propagation. But this capability, like the rest of OpticStudio, is really designed to handle traditional optics simulation. I know of no way to simulate in OpticStudio a horn antenna being modeled as an expanding waveguide.

Kind regards,

David