Implementation of Berenger layers as boundary conditions for the beam propagation method: applications to integrated waveguides

In order to numerically solve the wave equation in integrated optics device s, the implementation of appropriate boundary conditions should be a key is sue. In this work, we formulate the perfectly matched layer boundary condit ion, introduced by Berenger [J. Comput. Phys. 114 (1994) 185], for operatio n in the scalar bidimensional finite difference beam propagation method (FD -BPM). With this boundary condition, a high absorption of outgoing waves al ong the propagation direction is achieved. To test the operation of Berenge r layers, two numerical experiments have been performed: the propagation of a Gaussian beam in vacuum and the propagation of a guided mode in a slab w aveguide. (C) 1999 Elsevier Science B.V. All rights reserved.