Abstract

A set of full-matrix recursion formulas for the $W\to S$ variant of the S-matrix algorithm is derived, which includes the recent results of some other authors as a subset. In addition, a special type of symmetry that is often found in the structure of coefficient matrices (W matrices) that appear in boundary-matching conditions is identified and fully exploited for the purpose of increasing computation efficiency. Two tables of floating-point operation (flop) counts for both the new $W\to S$ variant and the old $W\to t\to S$ variant of the S-matrix algorithm are given. Comparisons of flop counts show that in performing S-matrix recursions in the absence of the symmetry, it is more efficient to go directly from W matrices to S matrices. In the presence of the symmetry, however, using t matrices is equally and sometimes more advantageous, provided that the symmetry is utilized.

© 2003 Optical Society of America

Formulation and comparison of two recursive matrix algorithms for modeling layered diffraction gratings

Lifeng Li
J. Opt. Soc. Am. A 13(5) 1024-1035 (1996)

Enhanced R-matrix algorithms for multilayered diffraction gratings

Eng Leong Tan
Appl. Opt. 45(20) 4803-4809 (2006)

Note on formulation of the enhanced scattering- (transmittance-) matrix approach

Eng Leong Tan
J. Opt. Soc. Am. A 19(6) 1157-1161 (2002)

Bremmer series, R-matrix propagation algorithm, and numerical modeling of diffraction gratings

Lifeng Li
J. Opt. Soc. Am. A 11(11) 2829-2836 (1994)

Modified S-matrix algorithm for the aperiodic Fourier modal method in contrast-field formulation

Maxim Pisarenco, Joseph Maubach, Irwan Setija, and Robert Mattheij
J. Opt. Soc. Am. A 28(7) 1364-1371 (2011)