Resource Optimization Of UAV-Enabled Mobile Edge Computing Systems With User Cooperation

The exponential growth of mobile applications has driven advances in wireless communications and networking.However,mobile devices generally have limited computing resources,which limits mobile applications that require powerful computing power such as virtual reality.In this regard,Mobile Edge Computing(MEC)is considered as an effective technique to address the limitations of mobile devices by providing dense computing services at the edge of the network.By offloading the computing tasks of mobile devices to adjacent MEC servers,the computing energy consumption and response latency of mobile devices can be significantly reduced.In addition,Unmanned Aerial Vehicle(UAV)-enabled MEC systems have been widely studied due to its ease of deployment.On the one hand,MEC servers can be deployed on UAVs to provide computing services for computationally intensive devices in remote areas where there are insufficient ground-based MEC servers or no infrastructure to support MEC servers.On the other hand,UAVs can use their highspeed mobility to quickly approach terminals and provide computing services,thus can significantly improve communication performance.In addition,considering the combination of Non-orthogonal Multiple Access and Wireless Power Transfer(WPT)technologies with UAV-enabled MEC can provide better performance for the system.However,the distributed nature and interconnectedness of mobile devices are rarely considered in UAV systems,therefore,this thesis investigates resource optimization of UAV-enabled MEC systems as follows.First,a UAV-enabled mobile edge computing system with user cooperation is considered which consists of a UAV carrying a MEC server,ground far users and ground near users.The near user is in the coverage area of the UAV,while the far user is not in the coverage area of the UAV,so the tasks of the far user need to be offloaded to the UAV with the cooperation from the near user.Considering the objective of minimizing the weighted sum energy consumption of the UAV and the users while satisfying the user’s quality of service,a two-step iterative algorithm based on the joint optimization of the UAV’s trajectory,the amount of data relayed and offloaded by the ground user,and the computation frequency of the Central Processing Unit(CPU)for the ground user and the UAV is proposed.Simulation results show that the proposed scheme is able to obtain lower total energy consumption compared to other benchmark schemes.Second,this thesis considers a UAV-enabled mobile edge computing system with user cooperation and WPT,where the system includes a satellite computing node,a UAV computing node and several ground active users and ground idle users.The ground active users have a certain number of computational tasks to be processed,and the UAV carries an energy transmitter that provides energy and edge computing services to the ground users.In addition,the UAV can relay some of the user tasks to the satellite for executing.In order to fully utilize the energy collected by the idle user,the active user offloads the computational tasks to the dynamically associated idle user for executing.Considering the objective of maximizing the active users’ task-input bits under the limitation of the UAV energy consumption,a three-step iterative algorithm is proposed based on jointly optimizing the UAV’s trajectory,the amount of data offloaded by ground active users,the CPU computation frequency of ground users,the UAV and the satellite,and the user association variables.Simulation results show that the proposed scheme obtains more task-input bits compared to other benchmark schemes.