Analytical Approach to Wave Scattering in Waveguide Junction with Conducting Cylindrical Posts
Malgorzata Warecka, Rafal Lech, Piotr Kowalczyk · 2026
Read Analytical Approach to Wave Scattering in Waveguide Junction with Conducting Cylindrical Posts by Malgorzata Warecka, Rafal Lech, Piotr Kowalczyk (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-05638.txt.
Introduction
Malgorzata Warecka and 2 co-authors propose, in a 2026 paper, a new exact approach to wave scattering inside waveguide junctions containing conducting cylindrical posts. Waveguides channel high-frequency waves through hollow structures, and posts inside a junction scatter the waves in ways designers must predict accurately. Their method builds on mode matching, a technique that fits wave patterns together at boundaries.
Local projection functions improve the numerical conditioning of the calculations, making results more stable. The approach also shrinks the boundary region where such sums are needed, raising speed. Several cases, including multi-section microwave filters, test convergence and accuracy. Comparisons with finite element results from a commercial simulator validate the method for real-world of such parts design. The tests span several worked cases, not just one.
Catalog record
| CREATOR | Malgorzata Warecka, Rafal Lech, Piotr Kowalczyk |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 18 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-05638
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Analytical Approach to Wave Scattering in Waveguide Junction with Conducting Cylindrical Posts Thanks: This work was supported by the Gdansk University of Technology through the Excellence Initiative–Research University within the Argentum Triggering Research Grants under Grant no DEC-19/1/2023/IDUB/I3b/Ag. Malgorzata Warecka, Rafal Lech, Piotr Kowalczyk Abstract A new approach for analyzing waveguide junctions containing conductive cylindrical objects is proposed. The algorithm is based on mode matching technique using local projection functions, which improves the numerical conditioning of the problem. Moreover, the approach enhances computational efficiency by reducing the boundary where the numerical integration is required. Both convergence and accuracy of the method were tested by using several examples, including multi-section microwave filters, and validated through comparison with results obtained using the finite element method implemented in a commercial full-wave simulator. I Introduction Waveguide technology has been used for decades in terrestrial and satellite communications, radar systems, and other high-power applications. Recent years have seen increased interest and rapid development in space and satellite technologies. Consequently, the demands on tools that enable fast and efficient simulation of structures used in these systems have also been growing [ 1 , 2 , 3 , 4 ] . Commonly used discrete methods are highly versatile, which is the reason for their widespread implementation in commercial full-wave simulators [ 5 , 6 , 7 ] . However, the versatility of these methods comes at the cost of limited efficiency. As a consequence, software dedicated to specific structures or geometries is constantly being developed. In such cases, a hybrid approach [ 8 , 9 , 10 ] or sometimes even analytical methods [ 11 , 12 ] can be used. This paper analyzes the analytical description of the electromagnetic wave scattering phenomenon in a rectangular waveguide junction containing cylindrical conducting objects. Similar structures have already been analyzed in the literature [ 8 , 11 , 12 ] . In all these cases, the analysis involved domain decomposition and introduced a hypothetical cylindrical region containing the investigated objects. The cylinder filled completely the junction (the diameter of the cylindrical region was equal to the width of the waveguide) and was connected to waveguide ports in which the incident and reflected waves were defined. The electromagnetic fields in the waveguide sections, typically expressed in Cartesian coordinates, were matched with the field in the cylindrical region, expressed as cylindrical functions. The projection used to fulfill the continuity conditions requires integration over the entire junction cross-section. In such cases, the integrals must have been evaluated numerically due to their complexity. Furthermore, the number of cylindrical functions within the junction and modes in the waveguides were closely related, leading to limitations on the number of modes, thus limiting the applicability of the method. In this paper an alternative approach is proposed, in which the junction region does not require the introduction of a hypothetical cylindrical region, and the numerical integration resulting from the projection of fields during matching is limited to the boundary of the scattering object. We use local basis functions known from one-dimensional finite element method (FEM) for this projection, since projecting global functions can lead to poor numerical conditioning. In this article, we analyze only conductive cylindrical objects and demonstrate the validity of the developed algorithm by comparing the obtained results with those obtained from the full-wave electromagnetic simulator InventSim [ 5 ] . It should be noted that the presented technique can also be adapted to analyze non-conductive objects with geometries other than cylindrical, described, for example, by an impedance matrix defined on the cylindrical surface surrounding the object. This topic will be addressed in our future work. II Formulation The analysis concerns a conductive cylindrical post located in a rectangular waveguide junction. It is assumed that the cylinder extends over the full height b of the waveguide. Fig. 1 shows geometry of the structure where a center of the post of radius R is located at a distance h from the waveguide wall. Further analysis addresses only the TE m0 modes propagating in the junction. Fig. 1: Geometry of the considered waveguide junction To determine the scattering matrix of the section, it is necessary to uniquely define the structure’s ports. We assume that the junction length is zero, and conventionally, port \mathrm{I} is located at z=0 plane, while port \mathrm{II} at z^{\prime}=0 plane. The fact that the cylindrical object is not entirely contained within the junction (section length equals zero) does not affect the validity of the analysis as shown for larger hypothetical cylinders [ 8 , 13 ] . The analysis requires defining the electric and magnetic fields in the cross-sections for both ports: E^{I}_{y}(x,z)=\sum^{M}_{m=1}p_{m}G_{m}(a^{I}_{m}e^{-\gamma_{m}z}+b^{I}_{m}e^{\gamma_{m}z})\sin(p_{m}x) (1) E^{II}_{y}(x^{\prime},z^{\prime})=\sum^{M}_{m=1}p_{m}G_{m}(a^{II}_{m}e^{-\gamma_{m}z^{\prime}}+b^{II}_{m}e^{\gamma_{m}z^{\prime}})\sin(p_{m}x^{\prime}) (2) H^{I}_{x}(x,z)=\sum^{M}_{m=1}p_{m}\frac{G_{m}}{Z_{m}}(b^{I}_{m}e^{\gamma_{m}z}-a^{I}_{m}e^{-\gamma_{m}z})\sin(p_{m}x) (3) H^{II}_{x}(x^{\prime},z^{\prime})=\sum^{M}_{m=1}p_{m}\frac{G_{m}}{Z_{m}}(b^{II}_{m}e^{\gamma_{m}z^{\prime}}-a^{II}_{m}e^{-\gamma_{m}z^{\prime}})\sin(p_{m}x^{\prime}) (4) where the parameters p_{m}=m\pi/a , \gamma_{m}=\sqrt{p^{2}_{m}-\omega^{2}\mu\varepsilon} , G_{m}=\sqrt{(2\mathrm{j}\omega\mu)/(\gamma_{m}abp^{2}_{m})} , Z_{m}=\mathrm{j}\omega\mu/\gamma_{m} depend on the waveguide medium. The coefficients a^{I}_{m} , b^{I}_{m} , a^{II}_{m} and b^{II}_{m} represent the incoming and outgoing signals at port \mathrm{I} and \mathrm{II} , respectively. The tangential components of the electric and magnetic fields in both ports must be continues on the segments L_{d}=\{x\in[0,h-R]\wedge z=0\} and L_{u}=\{x\in[h+R,a]\wedge z=0\} . In the case of boundaries with a conductor L_{c}^{I}=\{x(\varphi)=h+R\sin(\varphi)\wedge z(\varphi)=R\cos(\varphi),\textrm{ where }\varphi\in[\pi/2,3\pi/2]\} and L_{c}^{II}=\{x^{\prime}(\varphi)=a-h-R\sin(\varphi)\wedge z^{\prime}(\varphi)=-R\cos(\varphi),\textrm{ where }\varphi\in[-\pi/2,\pi/2]\} only the tangential component of the electric field is equal to zero: \begin{cases}H^{I}_{x}(x,0)=H^{II}_{x^{\prime}}(x^{\prime},0)\Big|_{L_{d}\cup L_{u}}\\ E^{I}_{y}(x,0)=E^{II}_{y^{\prime}}(x^{\prime},0)\Big|_{L_{d}\cup L_{u}}\\ E^{I}_{y}(x,z)\Big|_{L_{c1}}=0\\ E^{II}_{y^{\prime}}(x^{\prime},z^{\prime})\Big|_{L_{c2}}=0\end{cases} (5) The relations between the coordinates in the local coordinate systems used to describe the fields in each port are very simple: x^{\prime}=a-x , y^{\prime}=y , and z^{\prime}=-z . The orientation of the axes of the systems is as follows: \vec{i}_{x^{\prime}}=-\vec{i}_{x} , \vec{i}_{y^{\prime}}=\vec{i}_{y} , and \vec{i}_{z^{\prime}}=-\vec{i}_{z} . Fulfilling the continuity conditions ( 5 ) is traditionally achieved by projection using an appropriate set of functions. Unfortunately, global functions defined along the entire junction cross-section ( L_{d}\cup L_{c}^{I}\cup L_{u} ) or ( L_{d}\cup L_{c}^{II}\cup L_{u} ) are poorly conditioned numerically; as the number of functions increases, their linear independence deteriorates, leading to an ill-conditioned system. In the presented approach, we propose the use of local basis functions used in the one-dimensional (first-order) FEM. Their restricted domain improves numerical conditioning, and additionally, most integrations can be performed analytically. Each of the boundary segments ( L_{d} , L_{c}^{I} , L_{c}^{II} and L_{u} ) is uniformly discretized into subdomains \Omega_{k} , on which the following local functions are defined (see Fig. 2 ). Each of the local functions \alpha_{k}(t) is nonzero only on its assigned subdomain \Omega_{k} . The formula is as follows \displaystyle\!\!\!\!\!\!\!\!\!\alpha_{k}(t)=1-\frac{2|t-t_{k}|}{t_{k+1}-t_{k-1}},\quad t\in[t_{k-1},t_{k+1}]=\Omega_{k} (6) for internal regions of the domain where k=1,\dotsc,K-1 , while for the edges \displaystyle\!\!\!\!\!\!\!\!\!\!\!\!\alpha_{0}(t)=1-\frac{t-t_{0}}{t_{1}-t_{0}},\quad\qquad t\in[t_{0},t_{1}]=\Omega_{0} (7) \displaystyle\!\!\!\!\!\!\!\!\!\!\!\!\alpha_{K}(t)=1+\frac{t-t_{K}}{t_{K}-t_{K-1}},\quad t\in[t_{K-1},t_{K}]=\Omega_{K} (8) where t_{k}=t_{0}+k(t_{k}-t_{0})/K . The range and exact form of the function for each of the limit curves are specified in Table I . Fig. 2: Visualization of local basis functions on boundaries TABLE I: Local basis function parameters t t_{0} t_{K} K \Omega_{k} \alpha^{d}_{k} x 0 h-R K_{d} \Omega_{k}^{d} \alpha^{u}_{k} h+R a K_{u} \Omega_{k}^{u} \alpha^{cI}_{k} \varphi \frac{\pi}{2} \frac{3\pi}{2} K_{c1} \Omega_{k}^{cI} \alpha^{cII}_{k} -\frac{\pi}{2} \frac{\pi}{2} K_{c2} \Omega_{k}^{cII} Projection of the conditions ( 5 ) using the local functions defined above leads to a system of equations in which the only unknowns are the coefficients representing the incoming and outgoing signals at the ports \mathrm{I} and \mathrm{II} . We obtain 2K_{d} equations (for k=1,\dots,K_{d} ) \displaystyle\!\!\!\!\!\!\!\!\!\!\!\!\!\!\int_{L_{d}}H^{I}_{x}(x,0)\alpha^{d}_{k}(x)\;dx=\int_{L_{d}}H^{II}_{x^{\prime}}(x^{\prime},0)\alpha^{d}_{k}(x)\;dx (9) \displaystyle\!\!\!\!\!\!\!\!\!\!\!\!\!\!\int_{L_{d}}E^{I}_{y}(x,0)\alpha^{d}_{k}(x)\;dx=\int_{L_{d}}E^{II}_{y^{\prime}}(x^{\prime},0)\alpha^{d}_{k}(x)\;dx (10)