Prediction of Fading Phenomena for Link Budget Performance in Land Satellite Communication Links

Abstract

In this work, we create a unified model, which combines the pure statistical model, based on Markov stochastic process, and the physical-statistical model, based on statistical description of built -up terrain and theoretical description of physical shadowing effects caused by buil dings' overlay profile. Using such a combined model, we perform a theoretical prediction of outage probability of fading phenomena for land-satellite link budget design for mega-cell network performance. Results of theoretical predictions are compared with data obtained by the referred models and with precise experimental data, for which these models give best fit and correctly explain main propagation phenomena occur within the land-satellite links. A new simulation tool is constructed to predict link budg et in communication links for different satellite constellation schemes, for different satellite networks, different satellite elevation angles and operational frequencies. I. Introduction In the Land-satellite communication (LSC) link, local shadowing effects, caused by multiple diffraction from numerous wedges and corners of obstructions, become predominant and can significantly corrupt information sent from ground-based terminal to satellite and conversely. Below, we analyze two main concepts on how to account the terrain effects on LSC. First is based on statistical models, while another is based on physical-statistical models. To unify these models and to use them together in our analysis, we assumed, following assumption made in (Refs. 1, 2 ) based on Markov's chain model that the radio signal is moving within a channel only between two states: good and bad. Good state occurs when line-of-sight (LOS) component is predominant, while bad state occurs when LOS component is absent and only non-line-of-sight NLOS component is present due to shadowing and multipath phenomena. At the same time, we show below that the physical statistical models, which are based on classical aspects of radio propagation over the terrain and on the statistical aspects related to the statistical description of obstructions placed randomly at the rough terrain and of the random buildings' overlay profile, can also be converted in such a manner to use the same statistical parameters of Markov's chain, as it is done in pure statistical models. Therefore, we unify these approaches by using both statistical parameters of Markov's stochastic process, statistical distributions of terrain built -up features and propagation phenomena, to predict the link budget and the probability of fading for different satellite constellations elevation angles and conditions of the environment.