Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

 
Session Overview
Session
PB-A4: Wave propagation
Time:
Wednesday, 17/Jul/2019:
2:20pm - 4:10pm

Session Chair: Christophe Geuzaine
Session Chair: Yasushi Kanai
Location: Patio 44-55

Show help for 'Increase or decrease the abstract text size'
Presentations

Graphene Third-Order Electrodynamic Reponse Modeling via an Efficient Finite-Difference Time-Domain Technique

Stamatios A. Amanatiadis1, Nikolaos V. Kantartzis1, Theodoros T. Zygiridis2, Tadao Ohtani3, Yasushi Kanai4

1Aristotle University of Thessaloniki, Greece; 2University of Western Macedonia, Kozani GR-50131, Greece; 321-17-134, Omachi, Asahikawa 070-0841, Japan; 4Niigata Institute of Technology, Kashiwazaki 945-1195, Japan

The third-order electrodynamic response of graphene and the subsequently involved nonlinear effects are systematically investigated and numerically modeled in the present paper. Initially, the third-order conductivity of the two-dimensional material is thoroughly examined in order to transform it into a convenient expression for its efficient incorporation in explicit time-domain algorithms. Then, the simplified expression of graphene surface conductivity is straightforwardly embedded in an appropriately tailored 3D finite-difference time-domain method combined with an auxiliary differential equation scheme. Numerical examples, indicating the significant frequency upconversion, validate the proper third-order response modeling and accuracy of the featured technique.



Numerical Investigations of the Field Regions for Wire based antenna systems

Paul Baumgartner, Thomas Bauernfeind, Werner Renhart, Oszkar Biro

Graz University of Technology, Austria

The exact knowledge of the boundaries between the field regions around antennas is very important for e.g. measurement engineers or in the numerical simulation. Therefore, different definitions based on the varying needs can be found. In the presented work the quality of the boundaries suggested from the literature have been analyzed using an error estimation of different properties of the electrical field. The analyses are applied on dipoles with various lengths as well as dipole based arrays e.g. Yagi-Uda Antennas. In addition the attempt is made to derive a more convenient boundary between the field regions which better satisfies the IEEE standard. For the numerical computation the Method of Moments is applied.



Wiring Diagnosis using Time Domain Reflectometry and Random Forest

Mostafa Kamel Smail1, Yamine Sellami1, Houssem Bouchekara2, Abderahmane Boubezoul3

1Institut Polytechnique des Sciences Avancées, France; 2Department of Electrical Engineering - University of Hafr Al Batin, KSA; 3UPE-IFSTTAR/TS2/SIMU&MOTO, France

New wiring network diagnosis technique dedicated to embedded system based on time domain reflectometry response is proposed. The method is based on two complementary steps. The forward model allows to simulate the time response using RLCG circuit model and Finite Difference Time Domaine (FDTD) method, and to create the datasets. The inverse model allows to detect, localize and characterize the faults from the time response of the faulty network by using Random Forest techniqueTime Domain Reflectometry, Wiring Network, Fault diagnosis, Random Forest



An Anisotropic Polynomial-Chaos Technique for Assessing Uncertainties in Microwave Circuits

Christos I. Salis1, Theodoros T. Zygiridis1, Nikolaos V. Kantartzis2, Christos S. Antonopoulos2

1University of Western Macedonia, Greece; 2Aristotle University of Thessaloniki, Greece

This work presents a novel sparse polynomial chaos technique, which manages to efficiently extract the statistical moments of the S-parameters of a microwave filter. Specifically, a new anisotropic base index set is proposed that takes into account the influence of each random variable to the output quantity of interest. This scheme is combined with the D-optimal design of experiments, in order to produce reliable outcomes within low computational cost. Numerical results validate the efficiency of the algorithm.



A Hybrid Binary Grey Wolf Optimizer with Migration Operator for Discrete-Valued Antenna Design Problems

Sotirios K. Goudos1, Achilles D. Boursianis1, Nikolaos V. Kantartzis2, Traianos V. Yioultsis2, Christos S. Antonopoulos2

1Dept. of Physics, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece; 2Dept. of Electrical and Computer Engineering, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece

Several antenna design problems are discrete-valued and in order to be consistently addressed binary-coded algorithms, suited for combinatorial optimization scenarios, are required. Among them, grey wolf optimizer (GWO) is a recently proposed evolutionary technique based on grey wolves’ behavior. Although GWO has been originally designed for real-valued problems, discrete versions, also, exist. In this paper, a novel binary GWO algorithm, using a migration operator, like in the biogeography-based optimization (BBO), is presented. The hybrid methodology is applied to several discrete-valued electromagnetic problems, such as array thinning and arbitrary-shaped dual-band patch antennas. Results show that bGWO-BBO is very efficient and outperforms other popular binary algorithms.



Highly Parallelized Contour Integral Method for Computing Resonant Modes of Lossy Cavities

Vinh Pham, Wolfgang Ackermann, Herbert De Gersem

Darmstadt University of Technology, Germany

In this paper, we address an efficient solver of the Maxwell eigenvalue problem for lossy cavity resonators. The curl-curl equation for the electric field is discretized using curved quadratic finite edge elements resulting in a nonlinear eigenvalue formulation. The eigenvalue problem is efficiently solved using a contour integral method. The method enables an accurate computation of all eigenvalues within a pre-defined region and is implemented in a highly parallelized framework to enhance the performance of the algorithm. Numerical results are presented to demonstrate the accuracy and efficiency of the proposed method.



Study on the Stability of Perfectly Matched Layer Implementations Using a Hybrid Implicit-Explicit Update Scheme

Lilli Friederike Kuen, Rolf Schuhmann

Technische Universität Berlin, Germany

The Perfectly Matched Layer (PML) technique is a general purpose absorbing boundary condition for finite methods which is usually applied within explicit time-domain schemes. There are a number of slightly different implementations of the extended update equations, and many of them have been reported in literature to show some longterm instabilities, typically only weakly depending on the time step size. In this contribution, we present a hybrid implicit-explicit algorithm, that introduces two additional parameters to control the stability. The implicit part of the algorithm is used only locally within the PML region, and for a simple setup stability is reached for a certain range of the additional parameters.



PML Absorbers for Arbitrary Materials through General Vector Transformation Rules for Maxwell Equations

Bernard Kapidani, Joachim Schöberl

Institute for Analysis and Scientific Computing, TU Wien, Austria

A novel derivation of the Perfectly Matched Layer (PML) type of absorbing boundary condition is given based on the basic transformation rules for scalar and vector fields, which sheds light on the equivalence of the known approaches to complex stretching based PMLs for the solution of the time dependent Maxwell equations. Moreover, the general nature of this approach makes it naturally applicable to interfaces between PMLs and arbitrary materials (dispersive, conductive, bi-anisotropic). Numerical results within a Discontinuous Galerkin (DG) Finite Element Method (FEM) are given.



Spectral Domain Modeling of Ferrite-Loaded Device by Iterative GREEN’s Function Computation for Multilayer Structures

Fayçal BENMOHAMED1,3, Junwu Tao2, Didier Vincent1, Sami Youssef3, Noemen Ammar4

1Hubert Curien Laboratory, University of Jean Monnet, France; 2LAPLACE, Toulouse University, INP-ENSEEIHT, France; 3LPCMN Laboratory, Monastir University, Tunisia; 4SysCOM Laboratory, Tunis University, Tunisia

A spectral dyadic Green’s function of multilayer structure containing arbitrarily biased magnetic layer is developed. A systematic iterative procedure for structure with more than 3 layers is developed by numerical computation of transfer matrix. Applied to practical device such as differential ferrite loaded phase shifter good agreement is obtained between measured and computed results. The use of singularity extraction techniques allows to bypass the numerical truncation of Fourier transform and accelerate the convergence speed.



Electromagnetic physical numerical modeling of a GaN Distributed Transferred Electron Device based planar waveguide THz oscillator

Christophe Dalle

Institut d'Electronique, de Microélectronique et de Nanotechnologies, France

THE DEVELOPMENT, on a large scale, of THz applications remains still today dependent on the availability of solid-state THz power sources, namely powerful enough, low cost, compact and safely operating at ambient temperature. Among the possible electronic (transistor, varactor …) and optical (laser, quantum cascade laser …) solutions is the transferred electron device (TED). On the basis of the potential of the electron transport properties, gallium nitride (GaN) TEDs are expected to operate at THz frequencies. All the works previously devoted to TEDs have concerned lumped structures. An original and up to now never studied solution is the distributed TED (DTED) which present the advantage to be compatible with the monolithic integrated circuit technology. Thus, the purpose of this paper is to theoretically investigate the potential of GaN DTED oscillators as THz power sources. Because of the device highly non-linear RF behavior, depending on the local electron velocity/electric field interaction, the induced space-charge effects and the electron relaxation effects resulting from the high frequency operation, an accurate 2D time-domain numerical physical modeling has been developed. It is used to perform an exhaustive investigation of the oscillator RF operation based on the analysis of the space-time electromagnetic (EM) and electron transport physical quantities. This contribution presents some of the obtained results.



Efficient Scheme for Absorbing Boundary Condition of FDTD Method Based on Time-domain Surface Integral Equation

Hideki Kawaguchi

Muroran Institute of Technology, Japan

This paper discusses a method of absorbing boundary condition (ABC) by using a time-domain electric and magnetic integral equations (TD-EFIE/MFIE). Based on surface equivalence theorem, we can predict behaviors of electromagnetic fields at any positions and time outside a virtual surface S by using the TD-EFIE/MFIE on S. Accordingly, to adopt the predicted electromagnetic fields value to the outside boundary of the FDTD grid space, different type of ABC from the Mur's and PML ABC can be constructed. Then, the ABC using the TD-EFIE/MFIE is not necessary to be applied to location at a far field distance, and the FDTD grid space can be truncated at a near field distance. This means that effective memory saving is possible to use the ABC based on the TD-EFIE/MFIE. On the other hand, it is easily imagined that the calculation of the TD-EFIE/MFIE takes very long computation time. In this paper, a speed-up method for the TD-EFIE/MFIE is discussed by using a spherical harmonic expression of the TD-EFIE/MFIE.



An Estimation of Scattering Centers of Target using Multiple HRRPs

Kang-In Lee1, Sang-Hoon Jung2, Jong-Mann Kim3, Young-Seek Chung1

1Kwangwoon University, Korea, Republic of (South Korea); 2Seoul National University, Korea, Republic of (South Korea); 3Agency for Defense Development, Korea, Republic of (South Korea)

In this paper, we propose an estimation of scattering centers by applying the least square method (LSM) to high resolution range profile (HRRP) data measured at multiple observation angles. With this method, the computational complexity of the proposed algorithm can be greatly reduced by comparison with the conventional imaging method. In order to verify the performance of the proposed method, we compare the estimated scattering centers with those of the conventional imaging algorithm. Also, we show the computational complexity of the two methods.



Study of Statistical Characteristics of Channel Model with Ray Tracing Based Geometrical Optics method

Pengfei Lyu1,2,3, Julien Sarrazin3, Aziz Benlarbi-Delaï3, Xiaoyu Xu1,4, Shuai Yan1,4, Zhuoxiang Ren3,4

1Institute of Microelectronics, Chinese Academy of Sciences, Beijing, China; 2University of Chinese Academy of Sciences, Beijing, China; 3Sorbonne Université, Laboratoire d’Electronique et Electromagnétisme, L2E, 75005 Paris, France; 4Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, China

The high speed wireless communication with the carriers in the high frequency band is attracting much more attention in the recent years. The channel model for the high frequency in complex environment such as indoor communication is not well studied. In this paper, we propose to use the ray tracing based geometrical optics to study the influence of wave reflections on the channel model. The results of Monte Carlo simulation illustrates, to make the model manifesting the physical propagation, in particular for the time of arrival (TOA), the number of reflections should be considered as an additional parameter in the channel model.



Optimum Design of a Circular Patch Antenna with Metamaterial Substrate

Eduardo Souza1, Rose M. S. Batalha1, Thiago M. Machado-Coelho2, Gustavo L. Soares1, Elson J. Silva2

1Pontífica Universidade Federal de Minas Gerais - Pucminas, Brazil; 2Universidade Federal de Minas Gerais - UFMG, Brazil

This paper presents an optimization design for a circular patch antenna composed of metamaterial in its substrate. The genetic algorithm and particle swarm optimization are used to find the best design that aims to maximize the gain of the antenna. The optimization methods interacted with finite integration technique software to compute the antenna gain. Several experiments with different antenna configurations have been done. In this patch antenna design, a gain three times greater than the original version of the antenna was achieved.



A new application to discontinuous Galerkin time-domain method for the Surface Plasmons

Li Xu, Xing Li, Hao Wang, Zhong-Hai Yang, Bin Li

School of Electronic Science and Engineering University of Electronic Science and Technology of China, Chengdu, China

Nowadays, there are lots of research on surface plasmons (SPs) due to a series of novel electromagnetic characteristics, such as its remarkable near field enhancement effect, high sensitivity to dielectric environment. However, there are still great challenges for electromagnetic algorithms and commercial softwares to realize high precision and high efficiency electromagnetic modeling for the SPs devices. It is necessary to deal with the complicated electromagnetic response mechanism of the metal materials in the SPs devices. Especially, the multiscale model has an important effect on time schemes. Although the discontinuous Galerkin time domain (DGTD) method has many advantages now, the explicit time scheme is constrained by the stability condition due to small size of fine meshes. Therefore, we propose three methods to improve present drawbacks of the discontinuous Galerkin time domain (DGTD) method in this paper. Firstly, we introduce the Generalized Nonlocal Optical Response (GNOR) mode into the DGTD method, then a new explicit time scheme by combing the Lawson method with a low-storage Runge-Kutta (LSRK) is used, finally, non-conformal region decomposition is used to realize the efficient electromagnetic modeling of the SPs device. Some numerical results show that the methods are feasible.



An Enhanced Numerical Model to Represent Non-Homogeneous Media and Non-Linearities: a Grounding System Application Including Soil Ionization

Daniel S. Gazzana1, Alex B. Tronchoni1, Arturo S. Bretas2

1Federal University of Rio Grande do Sul (UFRGS), Brazil; 2University of Florida (UF), US

This paper proposes an enhanced numerical model based on Transmission Line Modeling method in two dimension (TLM-2D)

suitable for the representation of non-homogeneities and non-linarites of the medium. The methodology is focused on the evaluation of

the grounding system behavior including the non-linear soil disruption phenomenon. The novel algorithm considers the medium nonhomogeneity

by the properly inclusion of reactive and resistive stubs in the TLM circuit. Simultaneously, the soil ionization is

represented by the dynamic variation of conductive components present in model taking into account the residual resistivity retained

in ionized region. Simulations are presented considering a grounding grid submitted to a lightning surge.



 
Contact and Legal Notice · Contact Address:
Conference: Compumag 2019
Conference Software - ConfTool Pro 2.6.132+TC
© 2001 - 2020 by Dr. H. Weinreich, Hamburg, Germany