Estudo comparativo das técnicas para cálculo de atenuação devido à chuva

Abstract

Attenuation caused by rain can cause serious impact on communication services by radio that operate in frequencies over 10 GHz. For frequencies above this, more damage caused by rain occurs in satellite or terrestrial links. In general, to calculate with a degree of certainty the attenuation caused by rain in a certain locality, a history of precipitation is needed and more than this, the history must be long, that is, a period of registers of over 10 years of observation. The data in this study was obtained at the National Institute of Meteorology and from the book Drenagem Urbana published by Companhia de Tecnologia em Saneamento Ambiental de São Paulo. Information was obtained from this book concerning the intensity of rain which had a return prevision (when the rain can occur again). This information was of 2, 5, 10, 15, 20 and more years. The intensities of the return of rain of 2 years were converted to the equivalent in 1 year. With the data vector of 1 year, it is possible to arrive at that referent to 0,01% of this time, or 53 minutes, and apply it in the mathematical routines used to find the attenuations caused by rains in the selected cities. Data from ten cities was collected: Rio Branco (AC), Porto Velho (RO), Cuiabá (MT), Porto Alegre (RS), Curitiba (PR), Belo Horizonte (MG), Salvador (BA), Natal (RN), São Gonçalo (PB), and Quixeramobim (CE). These were chosen due to the fact that, as a group, they form a geographical corridor covering almost the totality of the Brazilian territory that is mentioned in the literature which refers to calculations of attenuation due to rain [3] and [4], with values of rain equal to 0,01% of the time either by the CCIR model that is 95 mm/h, or by the Crane model that is 67mm/h. Although this study does not intend to question the validity of these values in calculations, they marked the boundaries to compare with those obtained for each city, which are obviously different, not only because of the distance between them, but by the known annual rate in the regions. Even considering the fact that intensity rates of short rain tend to approximate for the several regions, the rates for the chosen cities are different from each other and these in turn are different from those recommended by the models mentioned. With the rain intensity of each city being known, it was possible, therefore to calculate the attenuations of signal using the models, obtaining the statistics and probabilities of occurrences either in satellite or terrestrial links.