Import Radar Cross-Section from Feko
Ray-optical simulations in WinProp can be accelerated if selected objects are replaced by scattering information that was imported from Feko.
A disadvantage is that many objects (many surface panels) may be needed to approximate an object (for example, a car) accurately enough. This can make the simulations time-consuming.
An alternative approach is to perform radar cross-section (RCS) simulations in Feko and use the results of such simulations in WinProp. An object in WinProp can then be simple; it can even be just a box. Whenever a ray hits the simple model, the RCS information from Feko is used to calculate the monostatic reflection. ProMan automatically corrects the (far-field) RCS to account for the finite distance, since in the near field, the wave fronts are not planar anymore.
To produce the necessary file with RCS information in a Feko simulation, set up a monostatic RCS simulation in Feko with an angular sweep of incident plane waves. Request far fields in Feko; select the option Calculate fields in plane wave incidence direction to obtain the monostatic RCS. On the Advanced tab of the far field request, select output to an ASCII file. The Feko solver will write a .ffe file to disk.
The location of the origin can be adjusted depending on the actual scenario of interest. In particular, it is often convenient to adjust the Z-coordinate of the RCS origin depending on the height of the radar system in the WinProp simulation.
- Theta is the angle with the Z-axis, for example, the vertical axis.
- Phi is the angle from the East in a counter-clockwise direction.
The azimuth angle can be adjusted in case the object, for example, the car, does not have the same orientation in the Feko and WinProp simulations.
An averaging filter is provided because the RCS results from Feko, at high frequencies, can vary by many dB within a couple of degrees. In the real world, where the radar system and other objects move in a complicated scenario, small and high reflections can alternate rapidly. Also, due to finite distances, deep nulls in the far-field RCS will be filled in. Therefore, the application of an averaging filter is appropriate to obtain more realistic results.