{Extension of Huygens principle to model dispersive media in computation in electromagnetics

UoM administered thesis: Phd

  • Authors:
  • Buraq Abdulkareem

Abstract

Electromagnetic simulators are essential tools for developing prototypes of devices such as microwave radios, radars, and biomedical engineering devices. These simulators mainly model the behaviour of the electromagnetic waves in inhomogeneous media, therefore, solving Maxwell's equations numerically is necessary, since they represent the propagation of the electromagnetic wave. Many methods have been introduced to solve Maxwell's equations; among them, Finite-Difference Time-Domain (FDTD) is the most widely-used method to solve Maxwell's equations in the time domain. The total field/scattered field (TF/SF) technique, which is based on the Huygens principle, is widely used in the FDTD method. The most common application for the TF/SF technique is the Huygens source, in which an incident wave is enforced in the computational domain. One drawback of the Huygens source is that, when the frequency approaches the cutoff frequency of the FDTD grid, a growing leakage is present in the scattered field region. Sophisticated methods have been introduced to overcome this issue in vacuum; in this thesis, two methods are proposed to implement the Huygens source in dispersive media such as Debye media. Besides the classical use of the TF/SF technique, it has been used in literature in other applications such as Absorbing Boundary Condition (ABC), named as HABC. The previous implementation of the HABC method depends on Higdon operator. In this thesis, we show that Higdon operator causes a high level of wave reflection if it is present in a non-vacuum medium. We introduce a new operator as an alternative to the Higdon operator in dispersive media. Furthermore, this operator is used to implement the HABC to offer an alternative ABC for the PML in dispersive media for limited bandwidth applications.

Details

Original languageEnglish
Awarding Institution
Supervisors/Advisors
Award date31 Dec 2017