SC334 The Art of Modeling Optical Systems

Tuesday, May 18, 2010
9:30 a.m.–12:30 p.m.
Curtis Menyuk; Univ. of Maryland, Baltimore County, USA
Level: Beginner (no background or minimal training is necessary to understand course material)


Course Description

This course will describe effective procedures for formulating and solving mathematical models of optical systems. A first step is to decide on the purpose of the model and the required accuracy: Design of a commercial system and prediction of a new phenomenon are typical applications, and the former requires more accuracy. You should then determine the basic equations and find analytical solutions when possible to serve as a baseline. You can then find computational solutions.

In obtaining a computational solution, you must first decide the problem's character. Is it a propagation problem, such as the nonlinear propagation of light in an optical fiber? Or is it a static problem, such as finding the modes of a semiconductor waveguide? Does it have a random component that requires Monte Carlo simulations? Once that is determined, you must choose between a number of different algorithms and determine what computer and what software to use. Whether using commercial or home-grown software, it should be validated in the parameter regime where it will be used. That can be done by comparison to analytical solutions when possible and by comparison to other simulation codes that solve the same problem. Verification by comparison to experiments is also important, but it is not the same as validation.

All these steps are not present in every modeling problem, but many of them typically are. In this course, the instructor will draw from his own experience in carrying out these tasks in modeling a variety of different optical systems to show how to do them.


Benefits and Learning Objectives

This course should enable you to:

  • Set up and test an optical system model.
  • Decide which software to use.
  • Analyze the time and space scales in an optical system.
  • Understand the different types of computational problems and how to solve them.

Intended Audience

Researchers who want to model their experiments and engineers who want to set up effective design models.


Biography

Curtis Menyuk received a doctorate from the University of California at Los Angeles in 1981, and he has worked at the University of Maryland College Park and SAIC. He is currently a professor of computer science and electrical engineering at the University of Maryland, Baltimore County. He previously co-directed a program in optical networking for the Department of Defense (1999–2001) and was chief scientist at PhotonEx Corp. (2001–2002). He specializes in modeling optical fiber and laser systems, and he is a Fellow of the OSA and the IEEE.