Utility-optimal load balancing and interference management in multi-cell networks

Abstract

To achieve high capacity, next-generation wireless cellular networks not only adopt advanced physical layer techniques and broader bandwidth but also consider the reuse factor close to one with the dense deployment of base stations (BSs) that cover small cells. Even in such networks, users at cell edges users would suffer from low throughput due to unbalanced user distributions among cells and severe inter-cell interference (ICI). Moreover, the scenario becomes worse in the heterogeneous cell structures, e.g., coexistence of pico/femto cells inside macro cells, because the portion of users whose capacity is limited by ICI grows. Thus, in multi-cell networks, the coordination of BSs to effectively manage ICI and resolve load imbalance is essential. In this thesis, various types of joint resource allocation problems for load balancing and interference management are considered in multi-cell networks. Basically, these are multi-dimensional problems in terms of the number of BSs/users, the degree of time/frequency/power and etc. To tackle such complex problems, we mathematically model the problems based on network utility maximization framework and propose optimal association, scheduling and power allocation algorithms. However, the potential gain from the joint multi-cell resources allocation also comes with several practical challenges such as large feedback overheads and high computations. To overcome these computational and signaling complexities, we also develop simple but efficient suboptimal algorithms. Rigorous analysis and extensive simulations demonstrate that our algorithms can not only closely approximate optimal algorithms but also practically implementable.