Research On Energy Efficiency Optimization Based On Resouce Allocation In Dense Small Cell Networks

With the rapid development of mobile Internet and communication technologies, a variety of applications have been emerged in an endless stream, resulting in the explosive growth of mobile data traffic. In the traditional macrocell networks, the demand for data traffic growth has been unable to meet. This challenge can be solved by heterogeneous networks where Small Cells are deployed in the macrocell networks. In addition,dense Small Cell networks, which consist of large numbers of Small Cells,have become one of the key technologies in 5G communication. However,deploying large numbers of Small Cells in the dense Small Cell networks will cause huge energy consumption. As one of the important metrics of wireless communication network, energy efficiency has attracted wide attention. Therefore, in order to realize the green communication of the future network, it is necessary to investigate the effective energy efficiency optimization algorithms to improve the energy efficiency and reduce the energy consumption of the dense Small Cell networks.The research of this thesis is sponsored by the Beijing Natrual Science Foundation Project "Energy efficiency network technologies research for heterogeneous cellular networks with coordinated Small Cells" (Grant No.4144079), and the Fundamental Research Funds for the Central Universities, named as “Interference management technology of Small Cell based on LTE-Advanced" (Grant No. 2013RC0111). In this paper, we focus on the energy efficiency problem in dense Small Cell networks through resource allocation technology. The main contents and achivements of the research are summarized as follows:1. The research contents of dense Small Cell networks are summarized. Firstly, the background of the introduction of Small Cell, the definition of Small Cell, the progress of standardization and deployment scenarios are introduced. Then, this thesis analyzes the challenges brought by the dense Small Cell networks, and summarizes the key technologies to solve these challenges. After that, the energy efficiency optimization of dense Small Cell networks is discussed in detail. The main energy efficiency metrics, the main approaches of energy efficiency optimization and the research status of these main approaches are reviewed. Finally, we summarize the challenges of energy efficiency optimization and point out the next research direction of this thesis.2. In order to address the problem that the energy efficiency optimization usually ignores the load difference between Small Cells in the dense Small Cell networks, a load-aware energy-efficient power allocation algorithm is proposed. First of all, the influence of base station load on energy efficiency optimization is analyzed, and the energy efficiency weighting function is designed. Then, the energy efficiency optimization problem is converted into a difference-of-conve programming problem,which is solved by the concave-convex procedure method. Finally, in order to reduce the signaling overhead, a simplified interference group-based load-aware energy efficiency optimization algorithm is proposed. The basic idea is that Small Cells form interference groups and each Small Cell only needs to exchange information with its interferers in each interference group. Simulation results show that the convergence of the proposed algorithms can be guaranteed, and the energy efficiency of the Small Cells can be optimized according to their loads. In addition, in the simplified group-based load-aware energy efficiency optimization algorithm, given the users' rate demands, there is an optimal size of the interference group,so that the energy efficiency of the Small Cell network can be the best.3. In order to solve the problem of poor performance of cell range extension (CRE) users caused by introducing CRE technology and low energy efficiency in the dense Small Cell networks, an energy-efficient joint frequency and power allocation algorithm is proposed, which can improve the network energy efficiency while optimizing the performance of CRE users. First of all, the frequency resource in the network is divided into two parts. One part is assigned to the CRE users of Small Cells and the central users of the macrocells, and the remaining part is assigned to the non-CRE users of Small Cell and the edge users of the macrocells. Then,algorithm based on non-cooperative game theory, macrocell power allocation is optimized to mitigate the cross-tier interference to CRE users.After that, based on graph coloring theory, the CRE users are divided into clusters, and the CRE users with severe interference between each other are divided into different clusters, and frequency resources are allocated orthogonally among clusters. Finally, by introducing interference pricing,Small Cell power allocation is optimized to mitigate the co-tier interference caused to CRE users and improve the energy efficiency of the Small Cell network. Simulation results show that the proposed algorithm can enhance the energy efficiency of the Small Cell networks and the throughput of the heterogeneous network. Meanwhile, the CRE users' performance can be significantly improved with slightly sacrificing the macrocell users'performance.4. Since the dense Small Cell network brings cooperation opportunities for the Small Cell base stations, in order to reduce the network energy consumption and improve the energy efficiency of the network, respectively, cooperative energy efficiency optimization algorithms based on subframe and power allocation are proposed. On the one hand, in order to reduce the network energy consumption while meeting users' rate demands, the cooperative sleep and power allocation(CSPA) algorithm is proposed. On the other hand, in order to improve the network energy efficiency while guaranteeing the throughput, a coalition-based sleep mode and power allocation (CSMPA) algorithm based on coalition game is proposed. In the CSPA and CSMPA algorithms, firstly,according to different optimization goals, we put forward the corresponding coalition formation condition, based on which, Small Cells form coalitions. In the same coalition, Small Cells are assigned to orthogonal activation subframes. Then, power allocation of Small Cells on each active subframes is optimized. In addition, the optimal number of activation subframes required by each Small Cell base station to meet the user's rate requirement is derived in the CSPA algorithm. Simulation results show that the CSPA algorithm can significantly reduce the network energy consumption and the CSMPA algorithm can improve the network energy efficiency without jeopardizing the network throughput. In both CSPA and CSMPA algorithms, more users' rate requirements can be guaranteed and Small Cells will not cooperate when the target rate reaches to a certain value.