Avaliação de transmissão de sinais M-QAM em modelos de canais sujeitos a ruído impulsivo e desvanecimento generalizado.

Abstract

This thesis presents a study of the transmission of digitally modulated signals in wireless communication systems subject to noise and fading. This communications channel model, unlike the previous ones, considers the channel from the reception point of view, whose main characteristic is the presence of local devices and equipment that generate impulsive noise, such as motors, switched sources, WiFi modems, among others. New fifth-generation communications networks and wireless sensor networks, which include IoT devices, are targets of this model, as are broadcast networks such as digital radio and television. The thesis presents a mathematical analysis of the combined effects of binary or multilevel double Gated Additive White Gaussian Noise (G2AWGN) models and η-μ, κ-μ or α-μ fading in the received signal. Exact and novel expressions for the calculation of the bit error probability (BEP) of the M-ary quadrature and amplitude modulation (M-QAM) scheme subject to G2AWGN noise and generalized fading models are determined. Variations of these expressions, obtained considering that the occurrence of impulsive noise is characterized with the aid of a random process C(t), controlled by a Poisson process (telegraphic process), are also presented. To analyze the results, PEB curves are presented as a function of the permanent signal-to-noise ratio for different values ​​of the impulsive signal-to-noise ratio, different probabilities of occurrences and durations of pulses and impulsive surges, constellation order M and different parameters that characterize the distribution of fading. A wireless communications channel model that applies Markov chains is also described in this thesis. The Markov chains better characterize the non-stationary channel, in relation to the classical models, because they represent a way to model sudden changes in the statistical behavior of variations in the intensity of the transmitted signal as the receiver travels through the communication environment, whose chain states can be defined according to the channel conditions for each considered scenario. Two diversity reception techniques are considered. Spatial diversity reception is called maximum ratio matching (MRC) reception, and polarization diversity reception uses quaternion algebra. These techniques are used to decrease bit error rates and make the communications channel less susceptible to the effects of impulsive noise and fading. New and exact PEB expressions, with the inclusion of these techniques, are presented in this thesis. A hypothesis test to determine, from observed samples, whether G2AWGN noise is present in wireless communications systems is also presented. Two approaches for estimating the parameters of the probability density function (PDF) that mathematically characterizes the G2AWGN noise, one using the method of moments (MOM) and the other using the maximum likelihood method, through the maximization of hope (EM) , are also considered. A strategy to mitigate the harmful effects of G2AWGN noise from the theory of optimal linear systems is also described. Numerical analyzes of the adopted mitigation strategy were performed for an OFDM signal model. For this model, it was found that the optimal linear filter designed can minimize the interference caused by G2AWGN noise.