Research On Task Offloading And Computing Resource Management In Mobile Edge Computing

Driven by the Internet of Things and mobile communication technologies,mobile terminals and compute-intensive applications are emerging.The performance of traditional mobile terminals can't meet the computing needs of computation-intensive and latency-sensitive applications.Mobile Edge Computing(MEC)technology has been widely concerned for its effectiveness in enhancing users' computing experience by sinking cloud resources(such as compute resources,storage resources,etc.)to the edge of the network.Efficient and reasonable allocation of edge cloud resources is important for achieving high reliability,low latency and high energy efficiency in computation offloading.Therefore,this thesis studied the method of task offloading and computational resource allocation in mobile edge networks.Firstly,a joint optimization scheme for task offloading and computational resource allocation in heterogeneous network scenarios is proposed,considering the impact of service diversity and network access heterogeneity on task offloading and computational resource allocation.Firstly,the joint task offloading and computational resource allocation is modeled as a mixed integer nonlinear programming problem based on Lyapunov queueing optimization theory,and the trade-off between offloading gain and delay is analyzed.Then a search-tree-based heuristic algorithm is proposed to optimize the task offloading and computational resource allocation strategy.In addition,an offloading priority criterion is designed for fast branch delimitation of the search tree.Finally,the simulation results verify the effectiveness and rationality of the proposed algorithm.Secondly,to address the optimization problem of edge cloud resources in a multiuser,multi-service,and multi-MEC server scenario,this thesis established a computational offloading and edge cloud computing resource allocation model for a multi-user-multi-service multi-MEC server scenario.In order to achieve efficient and reasonable allocation of server-side computing resources and to maximize system utility,an online computing resource auction algorithm(Multi-users,Multi-tasks,and Multiservers based online computing resources auction algorithm,M3RA)with offloading delay constraints is designed.Besides,the feasibility of the proposed M3 RA algorithm is demonstrated through computational complexity,individual rationality,truthfulness and budget balance.Finally,the effectiveness of the M3 RA algorithm is further proved by simulating the total utility of the system.