Geração de amostras do modelo de desvanecimento α-η-κ-μ e aplicações

Abstract

The transmission medium of signals in wireless communication systems is modeled through statistical distributions, which describe the random behavior of fading. There are several distributions models that are used for this purpose and which are consolidated in the literature. However, with the great revolution that is occurring in the world telecommunication scenario, more general and flexible statistical models are required. In this context, the recently proposed - - - distribution becomes attractive. This distribution takes into account the combination of the relevant short-term propagation phenomena and comprises many of the consolidated fading scenarios. In this dissertation, a revision of the fading model - - - and its main statistics are made due to the fact that this is a new proposal in the literature. In addition, the distributions - and - Extreme are also revisited, since the statistics of these models are of paramount importance for understanding the - - - distribution. It is proposed and analyzed a generator of random variables for the - distribution when 0 < < 0:5, making the generation of variables - with no restrictions. Furthermore, a simple and efficient algorithm is proposed for the generation of random variables - Extreme. Bearing in mind that the distribution - - - is the -th power of the sum of two independent - processes, a random variable generator with - - - distribution is proposed. In all proposed generators, the Kolmogorov-Smirnov test was performed to verify the quality of the sample generator. To prove the coherence and applicability of the - - - sample generation algorithm proposed, it is made the analysis in applications of wireless communication systems, namely, analysis of bit error probability, spectral sensing and systems with diversity.