Rain Impairment Model for Satellite Communication Link Design in South Africa using Neural Network

Over some decades, rain impairment has been considered as the major dominant impairment in microwave propagation especially at frequencies above 10 GHz. Tropospheric impairment gives rise to higher amount of loss, especially in the tropical and equatorial regions due to the high rain intensity in the region. Aside rain attenuation, other propagation impairments that affect satellite links include gaseous absorption, cloud attenuation and rain fade. These impairments usually posed some challenges in designing terrestrial and earth to satellite microwave links. In this study, the prediction of rain attenuation at frequencies above 12 GHz for line of sight earth to satellite links is investigated using 12 years (1994-2006) rain rate data for selected locations in South Africa. The prediction method use is based on machine learning ANNs (artificial neural networks) using rain rate and exceedance percentage as input data as a tool to predict rain-driven attenuation on the higher frequency satellite links in South Africa. This research aims to compare different models and make real time forecasting of rain attenuation data for earth-to space communication links (ESCL). The proposed model creditably observes good results for the Ka-down link frequency band in contrast to the ITU-R and Moupfouma models. The Results obtained will serves as a good tool for satellite-based digital transmission systems in South Africa