Joint Wireless Source Management And Task Offloading In Ultra-Dense Network

With the development of the 5th generation mobile networks(5G),a large number of delay-sensitive applications are continuously emerging,such as augmented reality,online interactive games,and Internet of Vehicles.However,the user terminal devices are limited by its own resources,which that cannot meet the delay and energy performance requirements of some applications,and requires high-performance computing machine to perform this applications.Ultra-Dense Network(UDN),as an important technology for achieving low latency in 5G communications,provides new ideas for this.By densely deploying small cell base stations,it can provide services for user terminal equipment at a closer distance and improve link transmission quality,thereby achieving the purpose of improving system capacity and Quality of Experience(Qo E).However,due to the limitation of wireless resources,how to improve the efficiency of computing task offloading is an important research content in UDN.For this reason,this article has conducted research on wireless resource management and task offloading in UDN.The main work contents are summarized as follows:(1)A heuristic algorithm for joint wireless resource management and task offloading based on Mobile Edge Computing(MEC)is proposed,which aims to minimize the time delay and energy consumption of task offloading.In the proposed heuristic offloading algorithm,the three wireless resources of MEC server computing power,wireless channel selection and user terminal equipment upload power control are jointly optimized.Considering that the researched joint wireless resource management and task offloading strategy problem is a Mixed-Integer Nonlinear Programming(MINLP)problem,the original problem is decoupled into the sub-problem of MEC server computing power allocation and wireless channel allocation and Upload power control sub-problem.The specific solution is as follows: for the problem of the optimal allocation of MEC server computing power,Lagrange function is used to solve the problem;for the problem of wireless channel allocation and upload power control,it is again decomposed into the problem of wireless channel allocation and upload power control,using greedy strategies and The golden section method is used to solve the problem.The final numerical simulation shows that the proposed strategy can meet the delay requirements of different user terminal equipment,so that the system has lower delay and energy consumption;and as the scale of user terminal equipment increases,the performance is better,which further verifies the proposed strategy.The algorithm is more suitable for ultra-dense network application scenarios.(2)A dynamic wireless resource management and task offloading strategy is proposed based on Lyapunov,where our objective is to minimize the average energy consumption of user terminal equipment under the stability condition of task queues.Lyapunov optimization technology is used to transform the long-term uncertainty problem into a deterministic problem in each time slot,and decouple the original problem into three subproblems: local computing power allocation,upload power control and offloading decisionmaking,and MEC server computing power allocation.The solution to the three sub-problems is described as: using convex optimization technology to solve the local computing power and upload power control;based on the user queue length to allocate MEC server computing power to offloading user terminal equipment;using the argmax function to evaluate the effective interference to obtain the offloading decision.The simulation results show that the proposed strategy shows superior performance in saving the energy consumption of user terminal equipment.