Several numerical algorithms, developed in the computational-fluid-dynamics community for solving the Euler equations, are found to be equally effective for solving the Maxwell equations in the time domain. The basic approach of these numerical procedures is to achieve the Riemann approximation to the time-dependent, three-dimensional problem in each spatial direction. The three-dimensional equations are then solved by a sequence of one-dimensional problems. This approach is referred to as a characteristic-based method. The basic algorithm can be implemented for both finite-difference and finite-volume procedures, and has the potential to eliminate the spurious-wave reflections from the numerical boundaries of the computational domain. The formulation and relative merit of the finite-difference and the finite-volume approximations are presented, together with numerical results from these procedures. >

Characteristic-based algorithms for solving the Maxwell equations in the time domain

A time domain unstructured grid approach to the simulation of electromagnetic scattering in piecewise homogeneous media

With the rapid development of the Internet of Things (IoT), it becomes convenient to use mobile devices to remotely monitor the physiological signals (e.g., Arrhythmia diseases) of patients with chronic diseases [e.g., cardiovascular diseases (CVDs)]. High classification accuracy of interpatient electrocardiograms is extremely important for diagnosing Arrhythmia. The Supraventricular ectopic beat (S) is especially difficult to be classified. It is often misclassified as Normal (N) or Ventricular ectopic beat (V). Class imbalance is another common and important problem in electronic health (eHealth), as abnormal samples (i.e., samples of specific diseases) are usually far less than normal samples. To solve these problems, we propose a sustainable deep learning-based heart beat classification system, called BeatClass. It contains three main components: two stacked bidirectional long short-term memory networks (Bi-LSTMs), called Rist and Morst, and a generative adversarial network (GAN), called MorphGAN. Rist first classifies the heartbeats into five common Arrhythmia classes. The heartbeats classified as S and V by Rist are further classified by Morst to improve the classification accuracy. MorphGAN is used to augment the morphological and contextual knowledge of heartbeats in infrequent classes. In the experiment, BeatClass is compared with several state-of-the-art works for interpatient arrhythmia classification. The <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> -scores of classifying N, S, and V heartbeats are 0.6%, 16.0%, and 1.8% higher than the best baseline method. The experimental result demonstrates that taking multiple classification models to improve classification results step-by-step may significantly improve the classification performance. We also evaluate the classification sustainability of BeatClass. Based on different physical signal data sets, a trained BeatClass can be updated to classify heartbeats with different sampling rates. Finally, an engineering application indicates that BeatClass can promote the sustainable development of IoT-based eHealth. 

BeatClass: A Sustainable ECG Classification System in IoT-Based eHealth

Time-domain local-operator methods have proven valuable in simulating electromagnetic phenomena from DC to light with simple and complex media. Certain limitations do exist with the time-domain local-operator methods including wall-clock simulation times and cells per wavelength requirements. This work achieves lower simulation times through code optimizations, algorithm optimizations and parallelism. This yields faster simulations times and lower cell per wavelength requirements. The improved method has been applied to a set of optical problems.

Higher-order-compact time-domain numerical simulation of optical waveguides

Monte Carlo method is a tool that has been widely used in materials science and engineering since its beginning in the decade of the 1940, but is also applicable in subjects as diverse as economics, society and its behavior, biology and even medicine. The working of the Monte Carlo method is based on the use of random numbers and to get to describe the behavior of a system or explain a phenomenon difficult to understand and to treat analytically. It aims to provide an introduction to the basics of Monte Carlo method applied to materials science, as well as show some examples based on the algorithm proposed by N. Metropolis and that has helped to address interesting problems in the field. At the same time basic examples are included with their own algorithms developed in the programming language FORTRAN 95 and illustrate quite well the foundations of the Monte Carlo method.

https://dialnet.unirioja.es/descarga/articulo/6684759.pdf

Fundamentos de simulación de materiales por medio del método de Monte Carlo

Accurate and rapid evaluation of radar signature for alternative aircraft/ store configurations would be of substantial benefit in the evolution of integrated designs that meet RCS requirements across the threat spectrum. Finite-volume time domain methods offer the possibility of modeling the whole aircraft, including penetrable regions and stores, at longer wavelengths on today’s supercomputers and at typical airborne radar wavelengths on the teraflop computers of tomorrow. To realize this potential, practical means must be developed for the rapid generation of grids on and around the aircraft, and numerical algorithms that maintain high order accuracy on such grids must be constructed.

Algorithmic Aspects and Supercomputing Trends in Computational Electromagnetics

Gigaflop (billion floating point operations per second) performance for computational electromagnetics

The ability to predict the electromagnetic response from complex structures with layered material media over a wide frequency range (100 MHz to 20 GHz) is a critical technology need for the development of stealth platforms. There are three basic approaches to numerical simulation of Maxwell’s equations: 1) high frequency asymptotics, which treats scattering and diffraction as local phenomena; 2) solution of an integral equation for radiating sources on (or inside) the scattering body, which couples all parts of the body through a multiple scattering process; and 3) the direct integration of the differential form of Maxwell’s equations in time. The integration of Maxwell’s equations in time offers the most direct and general solution for broadband radar scattering and propagation of electromagnetic pulses in real materials. The challenge for time-domain methods has been to maintain global accuracy in the phase and amplitude of waves scattered by large, complex structures.

Large-Scale Parallel Simulations in Computational Electromagnetics

The finite-volume time-domain method is applied to Maxwells equations in the time domain. A comprehensive and general-purpose computer code, CFDEM, written for this method was used to simulate the scattering of electromagnetic fields from the aft portion of a Navy ship. Computed results agree reasonably well with later measurements. >

Computation of the electromagnetic near field in the presence of a large structure

: Under sponsorship from various Department of Defense (DoD) organizations and a cooperative Research and Development Agreement (CRADA) with the US Research Development and Engineering Command, Army Research Laboratory (RDECOM-ARL) HyPerComp has significantly advanced the state of the art of time-domain, and broad band electromagnetic simulations The TEMPUS (Time-Domain ElectroMagnetic Parallel Unstructured Simulator) environment is a complete self contained code suite that includes computer aided design (CAD) geometry creation and repair unstructured gridding for fell-scale targets with general materials scalable parallel code architecture higher-order accurate discontinuous Galerkin solvers for Maxwell's equations and post processing utilities for solution visualization and exaction of final results like bistatic and monostatic radar cross section (RCS) synthetic aperture radar (SAR) images and high-range-resolution (HRR) profiles The high-performance TEMPUS environment is well suited for modeling a variety of targets and electromagnetic problems of interest to the US Army such as: 1) high- speed projectiles with subtle surface discontinuities ridges and/or fins 2) ground-based targets such as tanks and scud missile launchers and 3) foliage penetration and ground interaction for target under trees (TUT) Some of the physics-based phenomenological features that govern the electromagnetic response of general targets are: a) specular reflection b) creeping waves c) traveling waves) slow moving surface waves e) edge diffraction f) singular currents at surface discontinuities g) resonating gaps and cavities and h) general material response

Chris Rowell

Papers

Algorithmic Aspects and Supercomputing Trends in Computational Electromagnetics

Gigaflop (billion floating point operations per second) performance for computational electromagnetics

Large-Scale Parallel Simulations in Computational Electromagnetics

Computation of the electromagnetic near field in the presence of a large structure

Physics-Based High Performance Computing Using Higher-Order Methods for Broadband Applications in Computational Electromagnetics (CEM)