No abstract

No abstract

No abstract

In this work the Fuzzy Clustering technique was used to perform radio frequency signals prediction. This technique was used with georeferencing maps of topography and morphology for prediction of radio frequency power levels, at the region of Viçosa -MG. The performance of this method was evaluated through tests of propagation and mapping for a 879.660 MHz signal, used in cellular mobile telephony. This method of prediction showed excellent results in comparison with measurements of RF power levels with a success rate greater than the classical models of Okumura-Hata and Walfisch-Ikegami. Consequently, this method can be very useful to the telecommunications companies when making the RF cellular coverage prediction.

An alternative method is proposed to characterize the radiation pattern of antennas from an impulse response, based on a moment expansion together with a frequency domain near-to-farfield transformation via a FDTD/CPML solver. The approach introduced in the present work is intended for printed antenna analysis in special cases when the geometries to be modeled require a high degree of discretization in the space-time domain, which directly affects the frequency resolution when using FFT to perform the conventional NF-FF techniques. Results are compared with those obtained by a reference frequency-domain NF-FF technique and the applicability of the proposed method is demonstrated. isA deconvolution with less computational effort can be obtained using the expansion as in [14]