Research On System Energy Consumption And Transmission Strategies For UAV-Assisted Mobile Edge Computing

With the rapid development of 5G and Internet of Things technologies,many applications with a large amount of calculation and high latency requirements appear one after another,such as high-definition video processing,autonomous driving and real-time online games,etc.However,the computing resources of mobile terminal devices are often limited,and it is difficult to serve these emerging applications.Mobile Edge Computing(MEC)can place MEC servers at the edge of the network to provide computing support close to mobile terminal devices,so MEC is considered a reliable solution to the above problems.However,the traditional ground base station-based MEC system cannot effectively adapt to the dynamic computing needs of mobile terminals due to the fixed location of the base station.At this time,the Unmanned Aerial Vehicle(UAV)assisted MEC system has attracted the attention of the academic community.First of all,UAV can be deployed flexibly,so UAV carrying MEC servers can better approach mobile terminal devices to provide computing services.Secondly,since UAV fly at high altitudes,the communication link with mobile terminal devices is most likely to be a line of sight link,high communication quality,can effectively reduce communication delay.However,due to the limited onboard energy of UAV,it is very meaningful to study the energy consumption of UAV-assisted MEC systems.In addition,as the amount of data in the communication network continues to increase,the computing resources of a single UAV may not be sufficient to assist multiple ground users to complete computing tasks.At this time,the research on the system calculation amount of multi-UAVs assisted MEC is also worthy of attention.Base on the above reasons,this paper first study the problem of minimizing the system energy consumption of the UAV-assisted MEC,then the problem of maximizing the system calculation amount of multi-UAVs assisted MEC is studied.The main work of this paper is as follows:(1)Research on optimization of total system energy consumption of UAV assisted MEC system.In this system,considering that the calculation results of ground users can not only be used by themselves,but also may be shared with other users.Therefore,two scenarios are proposed: homologous UAV assisted MEC system(for its own use)and non-homologous UAV assisted MEC system(shared with other ground users).In the homologous scenario,the ground user offloads the computing task to the UAV,and the UAV returns the calculation result to the ground user itself.In non-homologous scenarios,ground users are divided into source ground users and destination ground users.Each source ground user corresponds to a destination ground user.The source ground user offloads the calculation task to the UAV,and the UAV returns the calculation result to the corresponding destination ground users.For these two scenarios,this paper proposes an optimization problem of jointly optimizing computing task division,UAV trajectory and slot length to minimize the energy consumption of the system,at the same time,the optimization goal can be changed to minimize the task completion time by changing the objective function.The problem is then solved using techniques such as Successive Convex Approximation(SCA).In particular,in the nonhomologous scenario,the initial trajectory of the UAV in this scenario has an important impact on the final result due to the causal relationship between each source-destination ground-user pair,therefore,this paper proposes to obtain the initial UAV trajectory by solving the Pickup-and-Delivery Problem.Finally,the simulation results prove that the proposed scheme can effectively reduce the energy consumption of the system.(2)Research on the optimization of system calculation amount of multi-UAVs assisted MEC.First,this system establishes a communication model between multiple UAVs and multiple ground users.Then under the UAVs airborne energy is limited,this paper proposes a joint optimization problem: optimize the scheduling between multiple UAVs and multiple ground users,the transmit power of ground users,and the trajectory of UAVs to maximize the amount of system computation.In order to prevent the extreme situation that some ground users have a very small amount of calculation,this paper introduces a fairness factor θ to ensure fairness.Then,multiple continuous variables are discretized by time discretization method,and the UAVs trajectory is initialized based on the traveling salesman problem.In order to solve this optimization problem,this paper first use Block Coordinate Descent to divide the optimization problem into three sub-problems.Each subproblem is then solved by introducing slack and using the SCA method.Finally,the detailed steps to solve the problem are summarized into an iterative optimization algorithm.Numerical simulation results proof that our proposed algorithm can converge quickly,and compared with the benchmark scheme,the target value of our proposed scheme is better.