ALGORITHM TO ESTIMATE DIFFRACTION LOSSES IN THICK IRREGULAR OBSTACLES

Abstract

Diffraction in obstacles is the physical phenomenon that predominates in those paths where the propagation of radiofrequency waves occurs without direct visibility between the transmitter and the receiver, and it is the fundamental cause of the basic propagation losses in these links. To quantify its influence, various methods are presented in the literature that vary in focus and complexity. These, based on the origin of their algorithms, are divided into three categories: the deterministic ones, which are based on mathematical expressions and which produce the same result under similar conditions, can be very exact, but when using integral solutions their computational implementation is extremely complex; The statistics, which are based on extensive measurement campaigns carried out under dissimilar reception conditions, their results implicitly bring a set of factors that consider elements of the environment that cannot be modeled analytically, but their accuracy depends on the similarities between the environment where will be applied and the one where it was validated; and the semi-empirical ones that use a combination of mathematical expressions and correction factors of empirical origin. The latter try to solve some deficiencies of the previous methods and currently are an adequate solution to solve the problem addressed. Semi-empirical solutions rely on the influence of geometry and the number of obstacles present in the path. Real obstacles can have very complex shapes, which justifies the use of approximate geometries (knife-edge) that generally do not consider their thickness, an element that undoubtedly influences the accuracy of the diffraction loss calculation. The aim of this article is to present an algorithm that considers the actual obstacle thickness during the calculation. The results obtained in the measurements carried out demonstrate the superiority of the algorithm in terms of accuracy, with respect to the approximate (semi-empirical) analytical solutions proposed in the specialized literature