Research On Optimization Strategy Of Computing Offloading Delay In Mobile Edge Computing

With the rapid development of Internet and 5G,the computing offloading research of computationally intensive and delay-sensitive tasks under mobile edge computing(MEC)has attracted extensive attention from the academic community.In this thesis,we study the delay optimization strategy of computing offloading in two scenarios of mobile edge computing,which is the scenario covered by multiple edge servers and the scenario of Mobile Edge Computing-Device to Device(MEC-D2D)assisted computing,respectively.We utilize artificial bee colony(ABC),bisection method,convex optimization theory and greedy algorithm in our thesis.The main work includes:In the scenario covered by multiple edge servers,we propose a task offloading and resource allocation algorithm considering fairness.Most of the existing research about mobile edge computing offloading focuses on the overall benefits of users,while few studies focus on the individual benefits of users,which results in the benefits of individual users being lower than that of most users.Therefore,it is necessary to ensure the benefit of each individual user through a reasonable task offloading scheme and resource allocation scheme.In this thesis,a joint model of task offloading and resource allocation is constructed for computing offloading in scenarios covered by multiple edge servers.Under the constraint of computing resources,communication resources of edge server and the maximum tolerable delay of tasks,we construct a model which aims at optimizing the benefits of all tasks under the premise of considering fairness.Then we propose a resource allocation algorithm based on bisection method and a task offloading algorithm based on artificial bee colony algorithm to solve the model.Finally,the simulation results indicate that our proposed algorithm performs better in the scenario of different task densities and different amount of resource.In the scenario of MEC-D2 D assisted computing,we propose a low-complexity algorithm of task offloading,task partition and resource allocation to maximize the overall user utility.In this thesis,a joint optimization model of task offloading,task partition and resource allocation is constructed for the computing offloading problem in the MEC-D2 D assisted computing scenario.To solve the model,a low-complexity joint optimization algorithm based on convex optimization theory and greedy algorithm is proposed.Finally,the simulation results show that the proposed algorithm has better performance in the scenario of different task densities,different D2 D equipment densities and different amount of resource.