چكيده به لاتين
Wireless communication channels are affected by destructive phenomena such as shadowing and multipath fading. Shadowing is relatively slow, so it is called large-scale effects, but multipath fading occurs due to the interference of constructive and destructive components caused by scattering, diffraction, reflection, etc., so it is called small-scale effects. Research in fading channel modeling is not a new issue and simultaneously with the introduction of wireless communication, many engineers and scientists have been involved in modeling communication channels. Therefore, so far, many models have been proposed according to the conditions and nature of the release environment.
In this thesis, we first have an overview of classic and new fading and shadowing models such as Rayleigh, Rice, Nakagami-m, Weibull, lognormal, gamma, generalized gamma, inverse gamma, α-μ, η-μ, κ-μ and α-η-κ-μ etc. Then, to evaluate the performance of these fading channels, different criteria such as amount of fading (AoF), outage probability (Pout), level crossing rate (LCR), average fading duration (AFD), bit or symbol error rate (SER/BER), channel capacity, effective rate and secrecy capacity are introduced.
In this thesis, four new distributions are proposed for fading channel modeling. Most of them are composite models. According to the propagation spectrum of new generation of communication systems, composite distributions are more appropriate models to describe the fading phenomena. The first proposed distribution is the α-μ/log normal model, which, as its name suggests, is a composite distribution. In this model, multipath fading has an α-μ distribution and shadowing has a log-normal distribution. The statistics of this distribution are calculated approximately based on Gauss-Hermite polynomials. The second model is the generalized Fisher distribution, which is a composite model, and unlike the Fisher distribution, which considers the distribution medium to be linear, adds the feature of the medium's non-linearity to this distribution. The third model is the NWDP distribution, which is a generalization of the TWDP. In this treatise, we first calculate the exact statistics for the TWDP distribution because the statistics of this distribution are approximate. We then extend this distribution and introduce the NWDP distribution where the number of specular components is arbitrary. The fourth distribution that is introduced in this thesis is the FNR distribution, which adds shadowing to the NWDP distribution, and in which the specular components fluctuate with the gamma distribution. After calculating the statistics of each of these distributions, different performance measures are computed for each one
