The market penetration of smart devices like smartphones and tablets with embedded communication technologies like WiFi, 3G and LTE has exploded in less than a decade. Complementing this technological trend, social networking applications have virtually connected a large portion of the population generating an ever-growing data traffic demand on the communication infrastructure. Pervasive communications have gained significance in the automobile industry as well, with the emergence of an impressive range of in-vehicle smart devices enabling driver assistance, infotainment, over-the-air vehicle monitoring, and even social connectivity on the move. This surge in the demand for connectivity has further challenged telecommuni- cation service providers to meet the expectations of high-speed network users. The goal of this thesis is to model and understand the mobility dynamics of high-speed users and their effect on wireless network architectures. Given the importance of developing our study on a realistic representation of vehicular mobility, we first survey the most popular approaches for the generation of synthetic road traffic and discuss the features of publicly available vehicular mobility datasets. Using original travel demand information of the population of a metropolitan area, detailed road network data and realistic microscopic driving models, we propose a novel state-of-art vehicular mobility dataset that closely mimics the real-world road traffic dynamics in both time and space. We then study the impact of such mobility dynamics from the perspective of wireless cellular network architecture in presence of a real-world base station deployment. In addition, by discussing the effects of vehicular mobility on autonomous network architecture, we hint at the opportunities for future heterogenous network paradigms. Motivated by the time-evolving mobility dynamics observed in our original dataset, we also propose an online approach to predict near-future macroscopic traffic flows. We analyze the parameters affecting the mobility prediction in an urban environment and unveil when and where network resource management is more crucial to accommodate the traffic generated by users onboard. Such studies unveil multiple opportunities in transportation management either for building new roads, installing electric charging points, or for designing intelligent traffic light systems, thereby contributing to urban planning.
