Study On Graph-based Interference And Resource Management In Heterogeneous Cellular Networks

The explosive growth of the various wireless services demands the further increase of the capacity of wireless network with the development of internet technology. The 3rd Generation Partnership Project(3GPP) proposes the heterogeneous cellular networks(HCN) in Long Term Evolution-Advanced(LTE-A)to increase the network capacity by spectrum reuse, where the HCN consists of macrocell and the overlapped low power nodes including the microcell, picocell, femtocell and relay. However, the joint management of interference and resource becomes an important issue. First, the resource allocations for multi-services are coupled due to the interference. Second, since the data rates on resource units are affected by the aggregation of interference and asymmetry of interference, to achieve higher accuracy of the influence of interference on rates with a low complexity is important for the fairness of users. Third, the multiple transmission patterns increase the complexity of the joint management of interference and resource. This dissertation mainly focuses on the joint management of interference and resource in HCN. We first address the resource allocation for multi-services in HCN. Then we study the fairness resource allocation of users under the interference control. At last, we investigate the resource allocation problem combing interference alignment.The dissertation is supported by the National Nature Science Foundation of China(No. 61231008), National Basic Research Program of China(No.2009CB320404), the 111 Project(No.B08038), Program for Changjiang Scholars and Innovative Research Team in University(No. IRT0852) and State Key Laboratory Foundation(ISN1002005, ISN090305). This dissertation studies the joint management of interference and resource in HCN scene. The main contributions of this dissertation are listed as follows:1. The resource allocation scheme for multi-service based on interference mitigation in HCN is studied. We divide the network into multiple meticulous interference regions by the minimum receival SINR threshold to mitigate the conflict reuse of resource with interference, which can improve the resource efficiency. On the one hand, we guarantee the performance of Guaranteed Bit Rate(GBR) services in macrocell that is not below the preset requirement. On the other hand, we give priority to the GBR services and improve the performance of non-GBR services to differentiate multi-service. We model our multi-service resource allocation problem as a multi-objective optimization problem.In order to solve the problem with a low complexity, we propose an algorithm based on the layering conflict graph to find the approximate solution, which is beneficial for our scheme to be applied in time-varying practical system. Simulation results confirm that the proposed scheme can increase the resource efficiency than the previous works and raise the satisfaction ratio of GBR services while improving the average performance of non-GBR services.2. The fair resource allocation scheme based on conflict graph with interference restricted in HCN is studied, which allows the resource reuse among interfered links with interference restricted. First, we model the fair resource allocation optimization problem with the physical interference model, which is non-deterministic polynomial hard(NP-hard). Then we propose a low complexity solution based on the conflict graph with interference restricted, where the proposed conflict graph considers the aggregate of interference and asymmetry of interference, which benefits the solution to approximate that in the physical interference model. Simulations results verify that the proposed scheme can improve the fairness of user effectively.3. The resource allocation scheme combing interference alignment(IA) base on the proposed transformed conflict graph is studied. In order to mitigate interference and improve the network performance, this part of the study proposes a resource allocation optimization problem combing IA to maximize the number of satisfactory users in Multiple-Input Multiple-Output(MIMO) femtocell networks. In the optimization problem, the mutual interferences inside the selected IA group are aligned into the interference subspace, while the rest of the interference in the network is mitigated by sub-channel assignment. To solve the problem with a low complexity, a three-phase solution framework based on graph theory is proposed. First, we give a transform rule to form the transformed conflict graph from the interference graph, while considering the mitigation of the mutual interferences among participant links and the occupation of the additional signal dimension of participant users in IA. Second, based on the analysis of the network topology, we give a selection criteria to select the feasible IA groups to increase the number of satisfactory users in the network. Third, based on the selection results of feasible IA groups and the transformed conflict graph, we give a sub-channel assignment scheme using the proposed modified maximum K-colorable subgraph algorithm. The simulation results show that the proposed scheme can approximate the optimal solution in small network, and improve the satisfactory ratio of users compared with the scheme without IA in dense femtocell network.4. The fair resource allocation scheme combing interference alignment is studied. Whether a user can perform IA or not brings about the unequal ability of interference mitigation. To improve the fairness of uses, we propose a resource allocation optimization problem combining the sub-channel assignment and IA to improve the fairness of users in MIMO femtocell networks, which is an NP-hard problem. In order to solve it with a low complexity, we propose a graph-based scheme. The selection criteria of IA groups is based on the analysis of the interference graph to improve the fairness of the users.