IEEE TRANSACTIONS ON MAGNETICS, VOL. 42, NO. 4, APRIL 2006 839 Numerical Modeling of an Indoor Wireless Environment for the Performance Evaluation of WLAN Systems Theodoros T. Zygiridis , Elissavet P. Kosmidou , Konstantinos P. Prokopidis , Nikolaos V. Kantartzis , Christos S. Antonopoulos , Konstantinos I. Petras , and Theodoros D. Tsiboukis Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, Thessaloniki GR-54124, Greece HELETEL, Hellenic Electronic Communications S.A., GR-15562 Holargos, Greece A site-specific numerical model, based on the finite-difference time-domain method, is developed in this paper for the indoor radio channel. The scenario of interest is concerned with wave propagation in a typical office environment, for which several simulations are performed considering different placements of the transmitting antenna. Both the 2- and 5-GHz bands are examined, where contempo- rary wireless local area networks operate. Important channel characteristics are evaluated via the estimation of received power levels, as well as the examination of small-scale fading and time dispersion. Index Terms—Finite-difference time-domain (FDTD) method, indoor propagation, wireless communications. I. INTRODUCTION T HE development and widespread utilization of wireless local area networks (WLANs) has triggered intense re- search concerning the study and prediction of indoor wave prop- agation. This is due to the fact that the understanding of the particular radio channel characteristics is considered essential for designing and realizing efficient wireless communication systems. However, in-building propagation is a highly complex process, as it occurs within environments possessing a variety of geometric and electromagnetic properties. Due to multiple in- teractions with walls, furniture, equipment, and people, received signals exhibit fading characteristics in space and spreading in time [1]. For site-specific calculations, deterministic approaches based on physical models that take into account all or most of the environment’s details constitute a reliable choice. In this frame- work, ray-tracing algorithms have been used extensively to per- form indoor wireless studies [2]–[4]. On the other hand, there also exist some applications of the finite-difference time-domain (FDTD) method [5] for similar purposes. For example, the im- pact of different wall models on several channel parameters was pointed out via FDTD simulations in [6]. Moreover, FDTD re- sults related to window stwere incorporated in a ray-tracing al- gorithm in [7], while [8] introduced a hybrid ray-tracing FDTD technique for accurate site-specific calculations. This paper presents a set of numerical simulations regarding WLAN operation in a typical three-room office, based on a two-dimensional (2-D) FDTD approach. To incorporate contemporary systems, both the 2.44- and 5-GHz bands are considered, with the latter examined separately at 5.25 and 5.8 GHz. Via the second-order accurate solution of Maxwell’s equations, field distributions are predicted with high spatial resolution. Based on the received signal strengths, power-cov- erage maps, as well as descriptions of local-field variations due Digital Object Identifier 10.1109/TMAG.2006.871649 Fig. 1. Layout of the examined office area. Nine potential placements of the transmitting antenna are indicated. to multipath propagation, are provided. Statistical characteriza- tion of small-scale fading is then performed by computing the corresponding Rician factors. Further wide-band simulations permit the study of the channel’s time dispersion, by analyzing the recorded power delay profiles (PDPs). In essence, the nu- merical results are exploited to assess fundamental properties of the examined wireless channel. Therefore, the proposed prediction tool can be practically utilized at the design stage of WLANs to ensure reliable services, without resorting to time-consuming or expensive measurement campaigns. II. SITE DESCRIPTION AND THE FDTD MODEL The indoor environment where WLAN operation will be investigated is a typical office area comprising three rooms, as depicted in Fig. 1. The examined region covers a space of 9.11 8.45 m. The walls are 20 cm thick, with conductivity 0018-9464/$20.00 © 2006 IEEE