| The introduction of new schemes that are based on the communication among distributed ports or nodes, has motivated the use of composite fading models due to the fact that the distributed ports/nodes experience different multipath fading and shadowing statistics, which will subsequently determine the required statistics for the performance analysis of different trans-receivers. However, most of the existing work on the performance of distributed antenna systems (DASs) in cellular networks rely on lognormal shadowing models, which do not lead to closed-form expressions for the composite fading statistics and for subsequent performance metrics. In this thesis, the more tractable generalized-K composite fading model is adopted. It is shown, using the moment matching method, that the generalized-K distribution can be region-wise approximated by the familiar Gamma distribution and then the statistics for both the sum and the weighted sum of generalized- K random variables are studied for both the independent and correlated scenarios.;Another line of research that was motivated by the introduction of DASs is the effect of the individual power constraints at each distributed antenna port (DAP) on the design and analysis of optimal/suboptimal transmission strategies. In this thesis, the optimality range of beamforming under the individual power constraints is derived for correlated two-input multi-output channels. Finally, the developed results on the statistics for the weighted sum of generalized- K random variables are utilized to study the performance of DASs in both single-cell and multi-cell environments to reveal the gains of cellular DASs and to develop insights into their design. |
