Mobile Edge Computing And UAV Relay Of System Application

The wide application of mobile devices and the rapid growth of Internet traffic have promoted the rapid development of mobile cloud computing(MCC).But it cannot meet the high requirements of ultra-high network bandwidth and speed,so Mobile Edge Computing(MEC)has been proposed in this case.However,traditional terrestrial MEC systems have certain limitations due to their high dependence on terrestrial infrastructure.While MEC networks that support unmanned aerial vehicles(UAV)can solve these challenges.Therefore,research on UAV-enabled MEC networks has important practical significance.This article has done the following research on the aspects of multi-user MEC systems,multi-user UAV-enabled MEC systems and practical feasibility schemes:First,this paper studies the energy optimization problem of computing offloading and resource allocation in the multi-user MEC system.We considered a multi-user MEC system,including a BS equipped with an MEC server and N single-antenna users.Each of users has one computing task.In this system,optimization variables such as joint optimization task offloading decision,transmission power,local computing resources and edge computing resources are considered,and then the task allocation optimization problem with minimizing user-side energy consumption as the objective function is studied.In order to solve this non-convex problem,we use the BCD algorithm to transform it into 4 sub-problems,and prove that the sub-problems are convex.We decide whether the user should offload the task or perform the task locally.Then we derive the closed expression of the optimal transmission power and the optimal local computing resource on the basis of determining the task offloading decision and finally we use Binary Search to obtain the optimal edge computing resources.Based on these steps,we propose a low-complexity joint algorithm for iteratively solving sub-problems and prove its convergence.The simulation results show that compared with other three algorithms,this algorithm consumes less energy.In addition,we study the optimization problem of minimizing the user-side time.Through joint optimization of task offloading decision,local computing resources and edge computing resources,the original problem is decomposed into three sub-problems,and the sub-problems are proved to be convex.By deciding whether the user should offload the task or perform the task locally,we derive the expression of the optimal local computing resource,and use the existing tools to obtain the optimal edge computing resource.Simulation results show that compared with other algorithms,this algorithm takes less time.Secondly,on the basis of the multi-user MEC system,a UAV is added,which mainly studies the user-side power problem in the multi-user UAV-enabled MEC system.The system has one UAV and K users.We consider four constraints: offloading decision,transmission power,local computing resources and UAV trajectory.In order to solve this non-convex problem,we use the BCD algorithm to transform it into four sub-problems,and prove that the sub-problems are convex problems.We propose a low-complexity joint algorithm for iteratively solving sub-problems and prove its convergence.The simulation results show that compared with other algorithms,this algorithm consumes less power.Finally,the problem of security capacity involved in real life of the multi-user UAV-enabled MEC system is studied.We consider a UAV-supported MEC system consisting of a UAV,multiple single-antenna UEs and an eavesdropper.In this system,the terminal can allocate computing tasks to the UAV in the presence of eavesdroppers.Specifically,through optimization variables such as the offloading decision of the joint optimization task,UE transmit power,UAV interference power and UAV flight trajectory,an optimization problem with the goal of maximizing security capacity is studied.In order to solve the non-convex problem,we use the BCD algorithm to decompose the problem into four sub-problems,and convert the four sub-problems into convex problems by using the SCA algorithm.Also we propose an iterative algorithm to solve the sub-problems,and further prove the convergence of the algorithm.The simulation results show that the method has good performance.