Semivectorial Helmholtz beam propagation by Lanczos reduction

Pao Lo Liu, Bing-Jing Li

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

By using the finite-difference method to perform Lanczos reduction, a semivectorial Helmholtz beam propagation algorithm is demonstrated. The applicability of this algorithm is no longer limited to paraxial beam and scalar-field. Mode indices of rib waveguides are calculated and compared to previously published data. Losses of Y-branch for two orthogonal polarizations are also presented. This algorithm is more efficient than the conventional beam propagation method in determining the mode index. To calculate the radiation loss, however, it requires much more computational effort. More than 30 Krylov vectors are needed to avoid numerical dissipation.

Original languageEnglish
Pages (from-to)2385-2389
Number of pages5
JournalIEEE Journal of Quantum Electronics
Volume29
Issue number8
DOIs
Publication statusPublished - 1993 Aug 1

Fingerprint

propagation
Beam propagation method
Finite difference method
Waveguides
dissipation
Polarization
scalars
waveguides
Radiation
polarization
radiation

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Physics and Astronomy (miscellaneous)
  • Condensed Matter Physics
  • Atomic and Molecular Physics, and Optics

Cite this

@article{33deaba252994ab69fe4e77777ddb373,
title = "Semivectorial Helmholtz beam propagation by Lanczos reduction",
abstract = "By using the finite-difference method to perform Lanczos reduction, a semivectorial Helmholtz beam propagation algorithm is demonstrated. The applicability of this algorithm is no longer limited to paraxial beam and scalar-field. Mode indices of rib waveguides are calculated and compared to previously published data. Losses of Y-branch for two orthogonal polarizations are also presented. This algorithm is more efficient than the conventional beam propagation method in determining the mode index. To calculate the radiation loss, however, it requires much more computational effort. More than 30 Krylov vectors are needed to avoid numerical dissipation.",
author = "Liu, {Pao Lo} and Bing-Jing Li",
year = "1993",
month = "8",
day = "1",
doi = "10.1109/3.245570",
language = "English",
volume = "29",
pages = "2385--2389",
journal = "IEEE Journal of Quantum Electronics",
issn = "0018-9197",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

Semivectorial Helmholtz beam propagation by Lanczos reduction. / Liu, Pao Lo; Li, Bing-Jing.

In: IEEE Journal of Quantum Electronics, Vol. 29, No. 8, 01.08.1993, p. 2385-2389.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Semivectorial Helmholtz beam propagation by Lanczos reduction

AU - Liu, Pao Lo

AU - Li, Bing-Jing

PY - 1993/8/1

Y1 - 1993/8/1

N2 - By using the finite-difference method to perform Lanczos reduction, a semivectorial Helmholtz beam propagation algorithm is demonstrated. The applicability of this algorithm is no longer limited to paraxial beam and scalar-field. Mode indices of rib waveguides are calculated and compared to previously published data. Losses of Y-branch for two orthogonal polarizations are also presented. This algorithm is more efficient than the conventional beam propagation method in determining the mode index. To calculate the radiation loss, however, it requires much more computational effort. More than 30 Krylov vectors are needed to avoid numerical dissipation.

AB - By using the finite-difference method to perform Lanczos reduction, a semivectorial Helmholtz beam propagation algorithm is demonstrated. The applicability of this algorithm is no longer limited to paraxial beam and scalar-field. Mode indices of rib waveguides are calculated and compared to previously published data. Losses of Y-branch for two orthogonal polarizations are also presented. This algorithm is more efficient than the conventional beam propagation method in determining the mode index. To calculate the radiation loss, however, it requires much more computational effort. More than 30 Krylov vectors are needed to avoid numerical dissipation.

UR - http://www.scopus.com/inward/record.url?scp=0027649729&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027649729&partnerID=8YFLogxK

U2 - 10.1109/3.245570

DO - 10.1109/3.245570

M3 - Article

AN - SCOPUS:0027649729

VL - 29

SP - 2385

EP - 2389

JO - IEEE Journal of Quantum Electronics

JF - IEEE Journal of Quantum Electronics

SN - 0018-9197

IS - 8

ER -