Advanced User-centric Modeling for Future Wireless Communication Networks: Performance Analysis and Optimization

Due to the increasingly growing demand for high data rates and a massive number of connected devices, future wireless communication networks are required to provide much more resources than the current networks can do. As an emerging solution for future cellular networks, dense deployment of small cell base stations (BSs) has received a great deal of attention both in academia and industry. A major challenge in dense cellular networks is the interference experienced by the user from its neighboring active BSs. The effect of such interference is more deleterious at cell-edge users which limits the density of deployed BSs. An effective promising solution is to move from a cell-centric to a user-centric paradigm which allows each user to be connected to a set (cluster) of BSs instead of being associated with a single one. This will mitigate the interference effect and remove the cell boundaries, i.e, no cell-edge users. In this thesis, we develop novel BS clustering models to enable a user-centric BS cooperation for future wireless networks. Unlike the existing clustering models, where a user is served by a cluster of BSs with fixed size (either a fixed number of BSs or fixed cluster radius), our proposed models adapt the cluster of each user dynamically based on its channel condition and quality-of-service (QoS) requirements. To design user-centric networks, we focus on several technologies introduced for future wireless wireless communication systems such as millimeter wave (mmWave) and terahertz (THz) networks, unmanned aerial vehicle (UAV)-assisted networks, hybrid multi-tier networks, and energy harvesting networks. We first investigate the performance of a user-centric mmWave network under the proposed dynamic BS clustering model using tools from stochastic geometry. To maximize the system spectral efficiency, an optimization framework for the user’s serving cluster is developed. Then, a user-centric THz system is designed to compensate for the high pathloss and hence improve the coverage of THz networks. Both dynamic and static clustering approaches are considered, based on which we study the coverage probability of the user-centric THz network by using stochastic geometry. Then, to design an energy-efficient and reliable air-to-air connection in UAV networks, we design a 3D user-centric clustering model where a set of UAV transmitters spatially distributed in a 3D space in the sky are carefully selected to serve another UAV receiver. Analytical expressions for the spectral efficiency and energy efficiency of this user-centric UAV network are provided and an efficient and tractable optimization framework to maximize its energy efficiency is developed. In this thesis, we also implement a user-centric BS clustering for hybrid networks where THz, mmWave, and sub6-GHz BSs coexist. In this system, a user can be associated with the best BS cluster, from either a sub6-GHz, mmWave or THz tier based on either the maximum SINR criterion or the maximum rate criterion. Thus, with carefully planned networks, enabling hybrid user-centric wireless systems can provide ultra-high rates while maintaining sufficient coverage in future multitier networks. Furthermore, we adopt the proposed user-centric clustering model to enhance the joint rate and energy coverage of cellular networks with simultaneous wireless information and power transfer (SWIPT). For this setup, we aim to insure that the user can harvest sufficient energy in a given time slot and receive the required minimum data from a given serving cluster. Then, a mathematical optimization model for the time switching coefficient is developed to maximize the system joint rate and energy coverage performance. All analytical results are validated by simulation with comparison to some of the existing works, demonstrating that the proposed analytical frameworks are accurate and efficient in the design and deployment of future user-centric wireless networks