A Study On BBU Computing Resource Allocation Algorithms For Mobile Cellular Networks

C-RAN(Cloud Radio Access Network),as a new type of wireless access network architecture,uses centralized base station processing and distributed RF transceivers to reduce the number of base stations and thus lower the energy consumption.Moreover,it uses the network virtualization technology and cloud computing mode to realize baseband processing resources sharing and reasonable allocation to increase the spectrum efficiency,and eliminate the "tidal effect" caused by the change of resource demands.But meanwhile,how to reasonably allocate the computing resources in the baseband processing pool to meet the needs of the RF transceivers in order to increase the utilization of the spectrum and reduce the energy consumption is a critical issue in the application of the C-RAN architecture.This thesis studies the allocation of the computing resources in a baseband processing pool in the C-RAN architecture.A couple of computing resource allocation algorithms are proposed to achieve optimized resource allocation for different network scenarios.First,the thsis introduces the concept and characteristics of the C-RAN architecture,and emphasizes on the network virtualization technology and cloud computing mode.Based on this,it introduces the problem of the computing resource allocation in the baseband processing pool.Then,the thesis considers the problem of minimizing the number of used BBUs for optimizing the BBU computing resource allocation and analyzes the performance for sovling the problem using four different algorithms,including the nextfit algorithm,firstfit algorithm,bestfit algorithm and genetic algorithm.Simulation results show that compared with the greedy-strategy based bestfit algorithm,firstfit algorithm and nextfit algorithm,the genetic algoritm can achieve the best performance but has a higer computing complexity.After that,the thesis considers the problem of balancing the used BBU computing resources under the premises of minimizing the number of used BBUs,and further proposes a firstfit algorithm and genetic algorithm based algorithm called HeuGA algorithm to solve the problem.Simulation results show that compared with the firstfit algorithm and the genetic algorithm,the HeuGA algorithm can achieve a better performance in minimizing the number of used BBUs and balancing the BBU computing resources.Finally,the thesis takes into account the additional computing resources caused by cross BBU computing resource demands,and aims to minimize the additional computing resources caused by cross BBU computing resources demands under the premises of minimizing the number of used BBUs and balancing the used BBU computing resources.An improved HeuGA algorithm is proposed to solve the considerd problem.Simulation results show that compared with the firstfit algorithm and the genetic algorithm,the improved HeuGA algorithm can achieve a better performance in minimizing the number of used BBUs,balancing the BBU computing resources,and minimizing the additional computing resources.