Three dimensional indoor electromagnetic wave propagation modeling using a hybrid method

Abstract: The objective of this project has been to develop a method to predict three dimensional indoor electromagnetic propagation.For this purpose, a hybrid method has been introduced, consisting in modeling the hallways as waveguides with Perfect Electric Conductor walls, and analyzing the propagation from a modal point of view. Two different methods were used in order to model changes in the guide, Modal Match- ing for changes in width and height, and Finite Differences in Time Domain (FDTD) for the cases in which the translation symmetry is not maintained. Cuboids were used for the shape of the FDTD sections. The Yee cell was used for  iscretization, and Uniaxial Perfectly Matched Layers to model windows and the different ports were the excitation is inserted, and the fields are extracted. This approach allows reflection-free ports, so they can be considered adapted, and therefore a modal scattering matrix representation becomes possible. The results for the Modal Matching subproblem were satisfactory, as opposed to those obtained by FDTD. For the case of a turn, the results for the mode TE10 were accept-able, despite presenting systematic errors and numerical artifacts, which got worse for the superior modes. The time requirements did not show a dramatic improvement, due to the number of matrix multiplications needed to sample the the fields in the ports located on the faces of the cuboid after every FDTD iteration. The conclusion obtained is that a new implementation of the FDTD subproblem is needed in order to achieve the objectives of accuracy and computing speed.