Research On C-RAN Based Energy-Efficient Resource Allocation Problems

The exponentially growing mobile data traffic and high speed data applications have driven the development of wireless access networks from 3G to 4G and toward 5G now,while also bring a lot of new challenges to the exist wireless network struc-ture,such as relatively low spectrum efficiency and high energy cost brought by the dense deployment of based stations.In the meantime,a new network structure named C-RAN was proposed by the industry and researchers provides new solutions to these problems.C-RAN is the abbreviation of Cloud Radio Access Network,it divides the baseband processing part and the radio part of traditional base station.The computa-tion resources for baseband processing in each BBU(Baseband Unit)are centralized in a baseband resource pool,connected with the RRHs(Remote Radio Heads)through highly reliable optical fibers.This kind of network structure can significantly lower the construction cost and some other corollary equipment cost such as the air conditioner and so on;on the other hand,the centralized processing structure also makes some large-scale interference coordination and resource allocation schemes possible in real wireless communication systems.However,the new structure also brings some new challenges to deal with,such as the requirement for new high efficiency multiple antenna cooper-ation algorithms and joint resource scheduling schemes.Under this background,this dissertation aims to research on the high energy-efficient resource allocation schemes in C-RAN.Firstly,the research is focused on the frequency resource allocation for OFDM(Or-thogonal Frequency Division Multiplexing)based C-RAN.The whole part is divided into two research points,the first one is the resource allocation scheme for traffic load in homogeneous cell deployment.Based on FFR(Fractional Frequency Reuse)scheme,each cell is divided into two zones,cell center zone and cell edge zone.A graph based dynamic resource allocation scheme is proposed for the traffic load in cell center zones with interference coordination among neighbor cells.Then with the cell center resource allocation results,the connection relationship between BBUs and RRHs can be estab-lished.With the application of the proposed scheme based on C-RAN structure,the utilization of frequency resource is improved.Also,the system energy consumption is reduced through reducing the number of BBUs in working model with higher system energy efficiency compare with the state of art.Then the dissertation expends the use of proposed graph based dynamic frequency resource allocation scheme to heterogeneous C-RAN with some changes to adapt to the new multi-level cell network.A system level simulation method is used to verify the performance of proposed scheme in heterogeneous network,and the results show that system energy efficiency is improved as well as the fairness for all users compare with the state of the art.Along with first research point,these parts of the whole dissertation provide resource allocation schemes and performance references for the evolution from our existing OFDM based wireless network structure to the new C-RAN structure.Secondly,the joint resource scheduling schemes is researched in the RRH cooper-ation based C-RAN with two kinds of traffic.One is the resource scheduling for static traffic jointly consider the RRH antenna resource and BBU processing resource.The coordinate transmission among different RRHs for one user is considered to improve the system throughput as well as suppress interference among different users.The disser-tation formulates the system energy efficiency optimization problem through the max-imization of proposed energy efficiency utility function.The optimization problem is divided into two subproblems,which are the network-wide beamforming optimization and BBU scheduling separately.The BBU scheduling problem is then reformulated as a bin-packing problem.The joint resource scheduling scheme can be achieved accord-ing to the solution of the formulated energy efficiency optimization problem,and the performances are verified through simulations.The other research point on RRH cooperation based C-RAN considers the delay-aware traffic.Based on the whole structure of C-RAN,the dissertation models the time varying traffic with queue model,and uses Lyapunov optimization method to formulate the optimization problem jointly consider the performance of user traffic delay and sys-tem energy efficiency.The optimization problem can be equivalently divided into three subproblems,which are the data access control,the BBU processing control and optimal beamforming strategy.The system long-term energy-efficient joint resource scheduling scheme can be achieved according to the solution of these problems.The theoretical and the simulation results show that the trade-off between system energy efficiency and user traffic delay can be achieved through the adjustment of a control parameter.These parts of research on RRH cooperation based C-RAN provide useful references for the design and application of next generation C-RAN based wireless network.The research results of this dissertation provide feasible solutions and schemes for the application of C-RAN in wireless network,which also gives important insights and practical application value for the design and deployment of next generation wireless network.