Mobile Small cells in Cellular Heterogeneous Networks

Abstract

The unprecedented soaring demand for capacity and coverage on cellular networks is challenging and straining operators. The current improvements in cellular standards are significantly behind the exponential growth in requirements. Cellular operators are currently shifting towards Heterogeneous Networks (HetNets) as the most promis- ing solution to meet user demands; by using a mix of Macro Base Stations (MBSs) and Small Base Stations (SBSs). Recently, several cellular operators have started outdoor deployments of small cells to enhance service in high-dense areas (e.g., downtown areas). In this the- sis we assess and propose HetNet solutions that capitalize on SBS deployments to boost capacity and coverage under varying scenarios. Initially we investigate the core challenge of SBS placement in high-demand outdoor zones. We propose dynamic placement strategies (DPS) for SBSs, and present two models that optimize placement while minimizing service delivery cost when feasibility is the core challenge, and minimizing macrocells utilization as their deployment, compared to small cells, pose a constant challenges. Both problems are formulated as Mixed Integer Linear Pro- grams (MILPs). These solutions are contrasted to two greedy schemes which we have presented and evaluated over extensive simulations. Our simulation results demonstrate that our proposed DPS achieve significant reductions of service delivery cost and MBSs utilization. Realizing that a significant cant amount of cellular demand is generated on the go and suffers deteriorating quality, recent research efforts proposed deploying SBSs onboard public transit vehicles to enhance cellular coverage. We investigate the potential performance gains of using mobile SBSs (mobSBSs). We assess and quantify the impact of utilizing mobSBSs which are deployed in vehicles to aggregate traffic and backhaul it to MBS. In our evaluation we study two important indicators to assess the Quality of Service (QoS) received by mobile users, and the ensuing network performance. Namely, we investigate Pairwise Error Probability (PEP) and Outage Probability (OP) for mobile users. Finally, we propose a novel mobile data offloading framework which capitalizes on mobile small cells and urban Wireless Fidelity (WiFi) zones to alleviate the data tra c load generated onboard on MBSs. We incorporate dedicated and adaptive offloading mechanisms that take into account mobile user service profiles (history) and WiFi coverage maps to improve the e efficiency of the offloading framework. We conduct extensive simulation experiments to evaluate the performance of the mobile offloading framework and contrast results to a benchmark.