Changes

Click here to learn more about the theory of [[SBR Method]].

[[EM.Cube]]'s SBR simulation engine can be used as a versatile and powerful asymptotic electromagnetic (EM) solver. If you compare [[EM.Cube]]'s [[Propagation Module]] with its other computational modules, you will notice a lot of similarities. While other modules group objects primarily by their material properties, [[Propagation Module]] categorizes the types of obstructing surfaces. Besides sharing the same ray-surface interaction mechanisms, all the objects belonging to a surface group also share the same material properties. [[Propagation Module]] offers similar source types and similar observable types as the other computational modules. For instance, the Hertzian dipole sources used in a SBR simulation are identical to those offered in PO, MoM3D and Planar modules. The plane wave sources are identical across all computational modules. [[Propagation Module]]'s sensor field planes, far field observables (either radiation patterns or RCS) and Huygens surfaces are all fully compatible with [[EM.Cube]]'s other computational modules.

As an asymptotic EM solver, the SBR engine can be used to model large-scale electromagnetic radiation and scattering problems. An example of this kind is radiation of simple or complex antennas in the presence of large scattering platforms. You have to keep in mind that by using an asymptotic technique in place of a full-wave method, you trade computational speed and lower memory requirements for modeling accuracy. In particular, the [[SBR Method|SBR method]] cannot take into account the electromagnetic coupling effects among nearby radiators or scatterers. However, when your scene spans thousands of wavelengths, an SBR simulation might often prove to be your sole practical solution.

[[EM.Cube]]'s new SBR simulation engine utilizes an intelligent ray tracing algorithm based on the concept of k-dimensional trees. A k-d tree is a space-partitioning data structure for organizing points in a k-dimensional space. k-d trees are particularly useful for searches that involve multidimensional search keys such as range searches and nearest neighbor searches. In a typical large radio propagation scene, there might be a large number of rays emanating from the transmitter that may never hit any obstacles. For example, upward-looking rays in an urban propagation scene quickly exit the computational domain. Rays that hit obstacles on their path, on the other hand, generate new reflected and transmitted rays. The k-d tree algorithm traces all these rays systematically in a very fast and efficient manner. Another major advantage of k-d trees is the fast processing of multi-transmitters scenes. Unlike the previous versions of the SBR solver which could handle one transmitter at a time and would superpose all the resulting rays at the end of the simulation, the new SBR shoots rays from all the transmitters at the same time.

In most scenes, the buildings and the ground or terrain can be assumed to be made of homogeneous materials. These are represented by their electrical properties such as permittivity e and electric conductivity s. More complex scenes may involve a multilayer ground or multilayer building walls. In such cases, one can no longer use the simple reflection or transmission coefficient formulas for homogeneous medium interfaces. [[EM.Cube]] calculates the reflection and transmission coefficients of multilayer structures as functions of incident angle, frequency and polarization and uses them at the respective specular points.

It is very important to keep in mind that SBR is an asymptotic electromagnetic analysis technique that is based on Geometrical Optics (GO) and the Uniform Theory of Diffraction (UTD). It is not a "full-wave" technique, and it does not solve Maxwell's equations directly or numerically. SBR makes a number of assumptions, chief among them, a very high operational frequency such that the length scales involved are much larger than the operating wavelength. Under this assumed regime, electromagnetic waves start to behave like optical rays. Virtually all the calculations in SBR are based on far field approximations.