Research On Joint Cloud And Wireless Resource Allocation Algorithms In Heterogeneous Networks

Recently,the applications demand of mobile terminals(MTs)in heterogeneous networks(HetNets)has showed a trend of great growth.Some applications need to be offloaded to the cloud server due to the restrictions on the battery capacity and computing power of MTs.Thus,the demand for wireless and cloud resources in HetNets increases significantly.Mobile edge computing(MEC)is given close attention for its cloud services with low-latency for MTs.In the HetNets scenario with MEC,this thesis studies the joint resource allocation algorithm based on the potential game and evolutionary game respectively to achieve the coordination management of the wireless and cloud resources effectively.The main contributions of this thesis are as follows.(1)This thesis introduces the advantages of HetNets,elaborates the research status and existing problems of resource optimization in HetNets with MEC.In addition,the application of game theory in resource optimization is introduced briefly.(2)A joint offloading and resource allocation(JORA)scheme based on potential game in HetNets with MEC is proposed.This study not only investigates offloading strategy,but also considers cloud and wireless resource allocation as well as interference management.Specially,the objective of the JORA scheme is to minimize the energy consumption and monetary cost while satisfying task completion time constraint from mobile terminals' perspective.The JORA problem is formulated as a distributed potential game and the existence of Nash equilibrium(NE)is proved.When tasks of MTs are offloaded to the cloud servers,a cloud and wireless resource allocation algorithm(CWRAA)is implemented to obtain subchannel allocation,uplink transmission power and the computation resource for the MTs.The solutions of subchannel allocation consist of uniform zero frequency reuse(UZFR)method and fractional frequency reuse method based on Hungarian and graph coloring(FFR-HGC).Simulation results show that the distributed JORA scheme can effectively decrease the energy consumption and task completion time with lower complexity.(3)A joint cloud and wireless resources allocation algorithm based on evolutionary game(JRAA-EG)for overlapping HetNets with MEC is proposed,where the objective is to satisfy mobile terminals'(MTs)offloading requirements and reduce MTs' cost.The cost function of JRAA-EG consists of energy consumption and time delay as well as monetary cost.The regions are divided according to the accessible service providers(SPs)and MTs in the same region form a population.An evolutionary game is formulated to model the SP selection and resource allocation of MTs.In addition,we prove the existence and uniqueness of evolutionary equilibrium(EE)in the evolutionary game model.The solutions of EE include centralized algorithm based on replicator dynamics and distributed algorithm based on Q-learning.Simulation results show that both algorithms can converge to the EE rapidly.Compared with existing schemes,the JRAA-EG can save more energy and have less time delay.(4)The current research work is summarized and the prospects of future research are given.