MoM antenna simulations, with Matlab: RWG basis functions

Abstract

More and more educators are using Matlab as a tool for teaching computational electromagnetics (CEM). The wealth of built-in functions and graphing tools makes Matlab very attractive for this (although other packages, such as Maple, also have their adherents). This paper discusses the implementation of a MOM code using the well-known RGW [Rao, Wilton, and Glisson] basis functions. The use of the Maflab PDE toolbox for mesh generation is discussed in some detail. Although primarily for educational purposes, the author's code also has applications in practical antenna analysis.The RWG basis functions used in this paperand very widely used in MOM analysisare now recognized as one of the first vector ("edge") elements to be used in CEM, although their connection to the Whitney (CTRN) elements, which revolutionized finite-element analysis around a decade back, was only appreciated fairly recently. (The RWG basis functions can be obtained by forming the cross-product of the normal vector and the CTLN hasis functions). This paper also includes a very useful list of references on the RWG functions.Abstract: The Matlab implementation of an MOM antenna simulation code is presented, primarily (but not exclusively) for educational purposes. The PDE toolbox is used to generate the mesh. Evaluating the MOM matrix entries using the Rao-Wilton-Glisson (RWG) basis functions is discussed. The delta-function generator is used to model the feed point. Built-in Maflab functions are used for post-processing. Results for several examples, including thin-strip dipoles, a monopole on a ground plane and a wide-band slot antenna, are presented.

Hybrid Tdie-Tdpo Method for Studying on Transient Responses of Some Wire and Surface Structures Illuminated by an Electromagnetic Pulse

“…2, the log periodic antenna consists of nine dipoles, the lengths of the longest and the shortest ones are chosen to be 0.345 m and 0.172 m, respectively. The structure parameters τ = 0.917 and σ = 0.07, the width of the PEC plate is 4 m, and the distance between the log periodic antenna and the PEC plate is 1.0 m. One delta-gap source is used to feed the antenna [30][31][32], the feed point is shown in Fig. 2(b) and the source is chosen to be a modulated Gaussian pulse.…”

Geometry Based Preconditioner for Radiation Problems Involving Wire and Surface Basis Functions

“…In those cases, it may be observed that the wire self-coupling dominance is emphasized as the frequency increases, which suggests the existence of a mesh-size dependence on the impedance matrix elements corresponding to the surface basis functions. That dependence is given by the surface basis common edge length of the RWG surface basis function, and it has been reported in [19] to account for the antenna impedance in small thin strip antennas. An alternative reading may be obtained by focusing the size dependence as a coupling imbalance between wire and surface subdomains.…”

Design and Analysis of Wideband Planar Monopole Antennas Using the Multilevel Fast Multipole Algorithm

“…As also depicted in Figure 4, the denominator in (1) accounts for the total current through the feed line of the monopole antenna. Unlike the base-driven model introduced for the modeling of monopole excitation [32,34,35], where a thin strip is used to the thin wire monopole or other kinds of antennas, the feed model in Figure 4 has exactly described the feed structure in real world, and the current flowing from the feed line to the radiation part of the monopole is contributed from all the surface current on the feed line. As compared with the base-driven model in [32,34,35], accuracy of the current computed at the feed point is greatly improved, because there is no approximation introduced for the feed structure in this new feed model.…”

“…Other simplified feed model such as the base-driven model [34,35], pin-feed model [36], and deltafunction generator model [37,38] are also widely used for the modeling of antenna excitations, but they are so simple that the input impedance would be inaccurate in some cases.…”