Research On Joint Computation Offloading Decision And Resource Allocation Algorithm In Mobile Edge Computing Network

In order to solve the contradiction between increasing computationally intensive tasks but limited computing capacity and battery capacity for mobile users,the contradiction between abundant cloud computing resources but limited access capacity,and the contradiction between significant growth in mobile broadband service investment and traffic and the gradual piping of bear-network and the average revenue reduction of users,mobile edge computing(MEC)technology was proposed.By deploying servers at the network entrance,mobile users are provided with highly reliable,low-latency computing and communication services.In the MEC system,mobile users can enhance their computing capacity,reduce the energy consumption by offloading computational intensive tasks to the edge server for execution.Meanwhile,eliminating the need to access the remote cloud to reduce backhaul link congestion and service response delays.MEC computation offloading couples of computing and communication.Therefore,user-aware performance depends on the joint allocation of computing and communication resources.In addition,resource allocation depends on users requesting computation offloading,so multi-user joint resource competition and offloading decisions also affect system and user performance.This paper aims to improve the performance of system and user offloading services,and studies the joint offloading decisions and resource allocation under the based on mobile edge cloud multi-user ultra-dense network and based on unbalanced edge cloud multi-user multi-task network.The main contents are summarized as follows:(1)Research on joint offloading decisions and resource allocation of ultra-dense network users based on mobile edge cloud.For MEC and ultra-dense network convergence scenarios,multiple base stations access the same edge server through a backhaul link.There are both users requesting computation offloading service and users requesting communication in the network.Each user requesting computation offloading service has single compute-intensive task to process,and the user requesting communication have the minimum rate requirement.We formulate the problem as joint channel allocation,power allocation,and computational resource allocation in order to minimize delay-power consumption weighted sum of users requesting computation offloading service under the minimum communication rate requirement.To solve the problem,we decompose the primal problem into a computational resource allocation problem and a joint channel allocation and power allocation problem.The computational resource problem using the KKT condition to obtain the optimal solution.For the joint channel allocation and power allocation problem,it is further decomposed into power allocation and channel allocation problems,and proposed two heuristic algorithms(CEP and ECEP)to obtain the suboptimal solution.The algorithm is simulated by changing the number of offloading users,the complexity of the task,the time delay weight,and the weight of the user.The simulation results show that ECEP outperform the CEP algorithm.(2)Research on the imbalance edge cloud based computing offloading for multiple mobile users(MUs)with multiple tasks per MU is studied.In which,several edge cloud servers(ECSs)are shared and accessed by multiple wireless access points(APs)with the backhaul links,and each MU has multiple computing intensive and latency critical tasks should be offloaded to execute,which involves both the AP association and ECS selection,with the objective of minimizing the offloading cost.Distinguished with existing research,besides the transmission delay and energy consumption from MU to AP,the ECS access-cost which characterizes the ECS access delay and(or)cost is introduced,thus finally formulates the delay-energy-cost tradeoff based offloading cost criteria.Both problems of minimizing the sum offloading costs for all MUs(efficiency-based)and minimizing the maximal offloading cost per MU(fairness-based)are discussed.And we proposed the centralized heuristic algorithms to obtain the suboptimal solutions.Numerical results are given at last to verify the efficiency and fairness of these proposed algorithms.