A random walk procedure for evaluating probability distribution functions in communication systems

Author

Yevick, David ; Bardyszewski, Witold

Author_Institution

Dept. of Phys., Univ. of Waterloo, Ont., Canada

Volume

16

Issue

1

fYear

2004

Firstpage

108

Lastpage

110

Abstract

We demonstrate that probability distribution functions (pdfs) such as those of interest in communication theory can be efficiently calculated through a random walk algorithm. To illustrate the procedure, we determine the pdf of polarization-mode dispersion in optical fibers. We find that the technique efficiently samples configurations with extremely low probability for which it converges far more rapidly than the multicanonical procedure.

Keywords

birefringence; convergence of numerical methods; error statistics; importance sampling; iterative methods; optical fibre communication; optical fibre dispersion; optical fibre polarisation; optical fibre theory; probability; random processes; Wang-Landau procedure; birefringence; communication systems; communication theory; convergence; histogram entry; importance sampling; iteration; optical fibers; polarization-mode dispersion; probability distribution functions; pulse distortion; random walk algorithm; scaling factor; stochastically varying vector; transition probability; Bit error rate; Communication systems; Histograms; Monte Carlo methods; Optical fiber communication; Optical fiber polarization; Optical fibers; Physics; Probability distribution; Sampling methods;

fLanguage

English

Journal_Title

Photonics Technology Letters, IEEE

Publisher

ieee

ISSN

1041-1135

Type

jour

DOI

10.1109/LPT.2003.818895

Filename

1255967