Abstract
The widespread data demand in emerging wireless cellular technologies necessitates the evolution of traditional networks’ deployment to accommodate the ever increasing coverage and capacity requirements. In emerging wireless systems a hierarchical multi-level network that consists of a mixture of outdoor small cells (relays) and indoor small cells (femtocell) deployments underneath the traditional macro-cell architecture can be seen as a key deployment strategy to meet these growing capacity demands. In such networks, Femtocell technology has attracted much attention as a key “player” to address coverage and capacity issues mainly in home and enterprise environments. However, a major challenge that arises in such indoor networks originates from the inter-cell interference between the femtocells (commonly known as co-tier interference), assuming that femtocells share the same spectrum. The main objectives of this thesis are to investigate inter-cell interference in femtocell networks and to propose efficient multi-cell scheduling mechanisms that can mitigate inter-cell interference in dense femtocell environments while maintaining spectral efficiency at acceptable level across the cells. We begin with investigating the impact of co-tier interference in femtocells, highlighting the necessity of interference mitigation mechanisms for arbitrary deployment of femtocells. In this direction, a novel low-complexity graph-coloring based interference coordination mechanism is proposed to be applied on top of intra-cell radio resource management. We additionally propose two locally centralized multi-cell scheduling frameworks that enclose adaptive graph-partitioning and weighted capacity maximization concepts. In particular, we decompose the problem in the latter case based on the Exact Generalized Travelling Salesman Problem as a close match in graph-based solutions. Extensive evaluation is provided by simulations showing a significant improvement over the state-of-the-art multi-cell scheduling benchmarks in terms of outage probability as well as user and cell throughput and thus the proposed algorithms are promising candidates of multi-cell scheduling in next generation small cell networks.