A data-driven modelng of large-scale mobile networks: Community and vehicular mobility

In this work, an overarching need of a framework for data-driven mobility modeling in pedestrian and vehicular networks is proposed. Mobility is one of the main factors affecting the design and performance of wireless networks.;In this work, first we focus on pedestrian mobility - we start our study by investigating the adequacy of existing mobility models in capturing various aspects of human mobility behavior as well as protocol performance. This is achieved systematically through the introduction of a multi-dimensional mobility metric space (MIDAS) to measure individual, pair-wise and group metrics. In addition, a methodical analysis of a range of mobile encounter-based networking protocols is conducted to compare the performance under various mobility models and extensive traces. Our results indicate significant gaps in several metric dimensions between real traces and existing models. We then introduce COBRA, a new mobility model capable of spanning the mobility metric space to match realistic traces. We further evaluate DTN protocols using various traces and models. Our findings clearly show that COBRA can match the realistic protocol performance, reducing the gap from 80% to 12%, and showing the efficacy of our approach based upon the metric space matching.;Realistic design and evaluation of vehicular mobility has been particularly challenging due to a lack of large-scale real-world measurements. To overcome these challenges, we introduce a novel framework for large-scale monitoring, analysis of vehicular traffic using freely available online webcams. So far, we collected 140 million vehicular mobility records from 4,909 cameras located in ten different regions. We represent cities by network graphs in which nodes are camera locations and edges are urban streets that connect the nodes. Such representation exhibits small world properties. Traffic densities show 80% temporal correlation during several hours of a day. It is found using the goodness-of-fit test that the vehicular density distribution follows heavy-tail distributions in over 90% of locations. Our analysis signifies that the traffic patterns are stochastically self-similar. We believe our work provide a much-needed contribution to the research community for realistic and data-driven design and evaluation of pedestrian and vehicular networks.