Maxwell's equations as a scattering passive linear system

George Weiss, Olof J. Staffans

Research output: Contribution to journalArticlepeer-review

24 Scopus citations

Abstract

We consider Maxwell's equations on a bounded domain Ω ⊂ 3 with Lipschitz boundary Γ, with boundary control and boundary observation. Relying on an abstract framework developed by us in an earlier paper, we define a scattering passive linear system that corresponds to Maxwell's equations and investigate its properties. The state of the system is (Formula Presentsd.) , where B and D are the magnetic and electric flux densities, and the state space of the system is X = E+E, where E = L2(Ω;3). We assume that Γ0 and Γ1 are disjoint, relatively open subsets of Γ such that (Formula Presented.). We consider Γ0 to be a superconductor, which means that on Γ0 the tangential component of the electric field is forced to be zero. The input and output space U consists of tangential vector fields of class L2 on Γ1. The input and output at any moment are suitable linear combinations of the tangential components of the electric and magnetic fields. The semigroup generator has the structure (Formula Presented.), where L = rot (with a suitable domain), Γ is the tangential component trace operator restricted to Γ1, R is a strictly positive pointwise multiplication operator on U (that can be chosen arbitrarily), and (Formula Presented.) is another strictly positive pointwise multiplication operator (acting on X). The operator -G is pointwise multiplication with the conductivity g ≥ 0 of the material in Ω. The system is scattering conservative iff g = 0.

Original languageEnglish
Pages (from-to)3722-3756
Number of pages35
JournalSIAM Journal on Control and Optimization
Volume51
Issue number5
DOIs
StatePublished - 2013

Keywords

  • Boundary control
  • Boundary observation
  • Maxwell's equations
  • Scattering conservative system
  • Scattering passive system

Fingerprint

Dive into the research topics of 'Maxwell's equations as a scattering passive linear system'. Together they form a unique fingerprint.

Cite this