Secrecy capacity maximization for a UAV-assisted MEC system

• Published in: China communications Vol. 17; no. 10; pp. 64 - 81
• Main Authors: Han, Dongsheng, Shi, Tianhao
• Format: Journal Article
• Language: English
• Published: China Institute of Communications 01.10.2020; Department of Electronic and Communication Engineering,North China Electric Power University,Baoding 071003,China
• Subjects: Delays; MEC; Optimization; physical layer secrecy; secrecy capacity; Servers; Task analysis; Trajectory; trajectory optimization; UAV communication; Unmanned aerial vehicles; Wireless communication; UAV communication; trajectory optimization; MEC; physical layer secrecy; secrecy capacity
• ISSN: 1673-5447
• DOI: 10.23919/JCC.2020.10.005

Unmanned aerial vehicle (UAV) communication has attracted wide attentions in the mobile edge computing (MEC) system owing to its high-flexibility and simple operation auxiliary communication mode. Users can offload computing tasks to UAVs, which serves as edge nodes. Meanwhile, UAVs forward the tasks onto a cloud center or base station for processing, thereby shortening the implementation time of tasks. Nevertheless, the offloading links of an UAV-assisted MEC system adopt a radio broadcasting mode. Several eavesdroppers might be present in the environment to eavesdrop the data sent by users and UAVs, thereby causing significant effects on the secrecy performance. An optimized iterative algorithm is proposed in this paper to realize the maximum secrecy capacity of the MEC system and further improve the secrecy performance of an UAV-assisted MEC system and assure secrecy transmit. By doing so, the secrecy transmit problems of the two-staged offloading model of the UAV-assisted MEC system are solved. The maximum secrecy capacity of the system is obtained through joint optimization of the UAV positions, transmit power of the UAV, task offloading ratio, and allocation of offloading users considering the limited time and energy of an UAV. Simulation results demonstrate that the proposed iterative algorithm can effectively improve the secrecy capacity of the system.