Stochastic geometry modeling and analysis for inter-cell interference coordination in 5G small cell networks

Abstract

Small Cell Network (SCN) is an inevitable response of the upcoming fifth generation (5G) mobile communication to explosively increasing demand for data traffic. By deploying small cells at high density in existing macro cellular networks, operators can serve users more closely, which improves spectral efficiency and capacity. However, inter-cell interference due to the full frequency reuse still limits the attainable performance of the network. As a solution to the challenge for inter-cell interference in SCNs, user-centric Inter-cell Interference Coordination (ICIC) has been drawing attention. On the other hand, in recent years more and more cellular network analyses have been based on stochastic geometry, especially homogeneous poisson point process (HPPP). This is because, unlike grid models and network-level simulations, HPPP captures well the characteristics of practical cellular networks such as irregular network topologies and gives mathematically tractable results. Modeling user locations and cellular networks based on HPPP is especially suited to evaluate the performance of user-centric ICIC since the model well reflects dynamic user locations and channel variations in cellular networks. In this context, several user-centric ICIC schemes have been proposed and analyzed based on HPPP, but there are some limitations. We focus on two limitations in the previous works. First, most of the previous works analyze their ICIC scheme applied to a typical user, not an edge user while the main purpose of ICIC is to improve the performance of an edge user. Second, most of the previous works do not consider intra cell resource allocation procedure in their ICIC scheme although the procedure plays an important role in user-centric ICIC. In this dissertation, we propose user-centric ICIC schemes to overcome these two limitations and develop a stochastic geometry model for performance analysis of our proposed ICIC schemes. In Chapter 3, we propose a distance-based ICIC scheme for SCNs. To accurately detect users experiencing severe performance degradation caused by inter-cell interference, we newly define an edge user in SCNs and then analyze the spatial distribution of edge users. Next, we apply our distance-based ICIC scheme only to the edge users, where base stations within a certain radius, called the cooperation radius, from each edge user cooperate to improve the performance of edge users. With the help of the stochastic geometry we obtain a semi-closed expression for the coverage probability of an edge user with the ICIC scheme. Based on our analysis we investigate two trade-offs on the resource efficiency of a network and the coverage probability of an interior user. We then determine the optimal cooperation radius that maximally improves the coverage probability of an edge user considering the two trade-offs. Our analytical results are validated through simulations. In Chapter 4, we propose an enhanced distance-based ICIC scheme for SCNs. To accurately detect edge users, we define an edge user in SCNs based on the average throughput degradation and then analyze the spatial distribution of edge users. Next, we apply our enhanced distance-based ICIC scheme to the edge users, where base stations give edge users higher weight than interior users so that frequency channels are allocated more frequently to edge users and base stations within a certain radius from each edge user cooperate to improve the average throughput of edge users. With the help of the stochastic geometry we obtain a semi-closed expression for the average throughput of an edge user with the ICIC scheme. Based on our analysis we investigate two trade-offs on the resource efficiency of a network and the average throughput of an interior user. We then evaluate the optimal average throughput of an edge user with the ICIC scheme considering the two trade-offs. Our analytical results are validated through simulations.