Absorbing boundary conditions (ABCs) for the finite-difference time-domain
(FDTD) method are introduced which arise from surrounding the simulation sp
ace with lossy Maxwellian material layers. Generalizations of the standard
Lorentz dispersion material model, the time-derivative and two-time-derivat
ive Lorentz material models, are developed for this purpose. The advantages
of this approach include the close connection of the ABCs with the actual
absorber physics associated with Maxwell's equations, the avoidance of the
field-equation splitting required by the Berenger PML layers, and reduced m
emory and operation counts. Several multi-dimensional cases are presented t
o quantify the efficacy of this Maxwellian material-based approach. (C) 199
9 Elsevier Science S.A. All rights reserved.