Physical–statistical methods for determining state transition probabilities in mobile‐satellite channel models

Abstract

Abstract Signal propagation in land mobile satellite (LMS) communication systems has for the last decade become an essential consideration, especially when high‐rate data services are involved. As far as urban or suburban built‐up areas are concerned, the extent of the influence is mainly restricted to the roadside obstacles, since the satellite is positioned at relatively high elevation angles in most practical situations. Probably, the most common model currently used for representing the LMS channel is the Lutz model, which uses two states to represent line‐of‐sight and non‐line‐of‐sight conditions. Transitions between these states are described by transition probabilities which are a function of the environment and the satellite elevation angles. Similarly, an extension to the model allows a four‐state description to be used for the states associated with a pair of satellites used in a dual‐diversity configuration. Calculation of the transition probabilities then requires knowledge of the correlation between the two channels, which in turn depends on the spatial characteristics of the local environment around the mobile. In both cases, the transition probabilities have been derived basically from measurements in the past. In the new approaches described in this paper, physical–statistical principles are applied to construct analytical formulas for the probabilities of shadowing and the correlation between states. These expressions apply particularly to systems operated in built‐up environments, and have been checked against numerical experiments and against direct measurements. In both cases excellent agreement is obtained. Copyright © 2001 John Wiley & Sons, Ltd.