Resource Allocation and Deployment Optimization for UAV-Assisted SPC-Based MEC Systems

We consider an unmanned aerial vehicle (UAV)assisted mobile edge computing (MEC) system, where a UAV-carried MEC server provides computation offloading service to multiple ground user equipment (UEs). To satisfy the low-latency requirement of the MEC, the short-packet communication (SPC) is used for the task offloading from the UEs to the UAV. The maximum computation latency among all the UEs is minimized by jointly optimizing the UAV’s deployment location, offloading bandwidths of all UEs, and computing times and CPU frequencies of the UEs and UAV. To solve the resulting challenging optimization problem, we propose an algorithm that decomposes it into two subproblems, which optimize the UAV’s deployment location, and the offloading bandwidths and CPU frequencies, respectively. In solving these subproblems, the data rate expression for SPC is approximated in a logarithmic form, and the equivalent problem reformulation and successive convex approximation techniques are used. Simulation results are presented to verify the effectiveness of the proposed algorithm in comparison with some benchmark schemes. Furthermore, the results show that the proposed algorithm is prominent for the system with SPC. based offloading, especially when the communication capability of the system is not strong.