Comportamento de modelos de cálculo de difração sobre terrenos irregulares nas faixas de VHF e UHF

• Ano de defesa: 2017
• Autor(a) principal: Vasconcelos, Lorenço Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Engenharia elétrica; Sistemas de comunicação móvel; Difração; Bullington; Deygout; Epstein-Peterson; Giovaneli; Gume de faca; Japonês; Terreno irregular; Diffraction; Irregular terrain; Japanese; knife-edge; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/18133; http://dx.doi.org/10.14393/ufu.di.2017.154

This paper compares five different methods for multiple knife-edge diffraction calculation on irregular terrain. Bullington, Epstein-Peterson, Japanese, Deygout and Giovaneli models were implemented in MATLAB® and the first four methods were compared with respect to the Giovaneli model because, in general, this model provides closer results to the rigorous Vogler solution. In addition, this paper presents the development of a software to calculate the attenuation of multiple knife-edge diffraction, electric field calculation and prediction of coverage area in the VHF and UHF bands on irregular terrain. The methodology is based on the Bullington, Epstein- Peterson, Japanese, Deygout and Giovaneli methods and on the use of topographic databases of terrain profile. It was found that the Bullington method provides the worst results and that Deygout model is often pessimistic. Moreover, Epstein- Peterson and Japanese models are very close to Giovaneli and can be used in many situations. These results provide an important base in the choice of methods to do diffraction analysis in practical radio links in the VHF and UHF bands. Finally, a correction is proposed for the Bullington model to make the results closer to the Giovaneli results, but without increasing the complexity of the method.