Abstract
M. Ing. (Electrical and Electronic Engineering)
This thesis investigates various methods of modeling fading communication channels.
These modeling methods include various techniques for the modeling of different fading mechanisms. The fading mechanism mainly considered throughout this thesis, is slow Rayleigh fading. Concise mathematical methods of modeling the effect that a fading communication channel has on binary digital data passing through it, is considered. A discussion on binary channel models, presented in the past, such as the Gilbert, Gilbert - Elliott and Fritchman channel models is also presented. All of these mathematical channel modeling techniques, are Markov chain based. Most of the investigations undertaken during the course of this work were based on Fritchman channel models. Results indicating the capabilities as well as the short falls of Fritchman models are presented. The experimental results presented here were obtained using dedicated error recording equipment, designed especially for the purpose of undertaking measurements, which enables one to deduce mathematical channel models. Detail regarding this equipment can be found in one of the two accompanying volumes, volume II. Furthermore, a new approach to channel modeling, based on hidden Markov models, is also presented for the modeling of channels with soft decision outputs. Results proving the accuracy of the proposed soft decision modeling technique, are also presented.