Joint Optimization of UAV Position, Time Slot Allocation, and Computation Task Partition in Multiuser Aerial Mobile-Edge Computing Systems

In this correspondence paper, an unmanned aerial vehicle (UAV) enabled mobile-edge computing (MEC) server is considered to provide MEC services for multiple ground users by employing a time-division multiple access protocol. Our objective is to solve the joint optimization problem of UAV position, time slot allocation, and computation task partition, which minimizes the system energy consumption of all users while ensuring the successful task computation of all users during a time block. We propose a globally optimal solution, which is found by two-dimensional search over possible UAV positions. In each search, we theoretically derive the semi-closed-form solution to the joint optimization of time slot allocation and computation task partition by employing augmented Lagrangian active set method. An alternating optimization scheme to find the locally optimal solution is also proposed. It is shown through numerical results that our proposed schemes are superior to the no offloading, full offloading, and gravity center offloading schemes.