Issue |
Eur. Phys. J. Appl. Phys.
Volume 57, Number 1, January 2012
|
|
---|---|---|
Article Number | 10501 | |
Number of page(s) | 5 | |
Section | Photonics | |
DOI | https://doi.org/10.1051/epjap/2011110113 | |
Published online | 02 December 2011 |
https://doi.org/10.1051/epjap/2011110113
Efficient perfect matched layer for hybrid MRTD-FDTD computation with lossy dielectric material boundaries
1
Group of Electronic and Telecommunications Systems, Universidad de Los Andes, Bogotá, Colombia
2
Lab-STICC, CNRS, Telecom Institute, Telecom Bretagne, Brest, France
a e-mail: g_massy@yahoo.com
Received:
17
March
2011
Revised:
25
September
2011
Accepted:
3
October
2011
Published online:
2
December
2011
Existing implementation of perfectly matched layers (PML) constraining the computational volume in an electromagnetic modeling of open structures has been shown to be efficient. However, cases with absorbing materials inserted into PML require substantial amount of memory and calculation time spent on the computation of the field inside the PML layer. In this paper we are addressing these problems by introducing a new computational method. As the procedure requires some substantial amount of memory and operations to compute fields in the PML layer, the paper addresses these issues. The PML split-field equations are revisited yielding a reduction from four-term storage to only three. In addition, the new algorithm is used with a new hybrid MRTD-FDTD (HMRTD) method applied to various lossy structures, conjointly with a higher-order field evaluation. Results demonstrate the efficiency of the new PML algorithm as compared to current schemes.
© EDP Sciences, 2011
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.