Distributed Optimization of Multi-Role UAV Functionality Switching and Trajectory for Security Task Offloading in UAV-Assisted MEC

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) have become a hotspot in recent years. However, due to the broadcast nature of wireless channels, it is difficult to avoid the presence of eavesdroppering in the networks, which can severely influence the users privacy protection between users and UAVs. In order to enhance the communication security of the networks while reducing system latency, this paper considers a dual UAV-assisted edge computing system that contains two UAVs: one UAV acts as an MEC server to process users' offloaded computation tasks, and the other multi-role UAV can function as a jammer or a relay, which depends on the users' expectations, to improve communication security and reduce system latency. However, due to the variables coupling and inconsistency in user preferences for UAV functionalities, the jointly optimization of UAV trajectories, computation resource allocation, and multi-role UAV's functionality switching is challenging. To address this issue, this paper proposes Distributed UAV Functionality Switching and Trajectory Design (DUFST) algorithm, which utilizes block coordinate descent (BCD) and successive convex approximation (SCA) to decouple the variables. Then, DUFST adopts the alternating direction method of multipliers (ADMM) to tackle the problem of inconsistent user preferences for UAV functionality selection. Simulation results verify the effectiveness of this algorithm, which outperforms other comparative algorithms in terms of performance.