Thermal-Aware Task Scheduling and Resource Allocation for UAV-and-Basestation Hybrid-Enabled MEC Networks

• Published in: IEEE transactions on green communications and networking Vol. 7; no. 2; p. 1
• Main Authors: Bao, Lingyan, He, Zhenli, Tan, Jinghong, Chen, Yu, Zhao, Mingxiong
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.06.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Central Processing Unit; Central processing units; Computing or caching cooperation; Computing time; CPU temperature; CPUs; Edge computing; Energy consumption; Hovering; hovering trajectory; Hybrid systems; Internet of Things; Mobile computing; mobile edge computing (MEC); Processor scheduling; Resource allocation; Resource management; Resource scheduling; Rural areas; Servers; Task analysis; Task scheduling; Travel time; Unmanned aerial vehicles
• ISSN: 2473-2400; 2473-2400
• DOI: 10.1109/TGCN.2023.3241954

Unmanned aerial vehicle (UAV)-enabled mobile edge computing (MEC) has recently emerged to provide services in the resource-shortage areas, e.g., rural and disaster rescue fields. However, to provide satisfactory services, especially for computation-intensive and time-sensitive tasks, UAVs are required to operate at their full capacity due to their limited computation resources. This may lead to excessive heat dissipation and damage computing units. To tackle this dilemma, this paper takes central processing unit (CPU) thermal constraints during UAV operating hours into consideration. To further relieve the burden on the UAV, we explore a hybrid MEC system with a base station and a hover-fly-hover UAV that is able to compute during flight. We aim to minimize the cruise duration of the UAV by jointly optimizing user admission policy, task scheduling, UAV hovering trajectory, traveling time, and computation-and-communication resource allocations. The formulated problem is complicated and we further divide it into four sub-problems which are alternating to be solved in sequence. Simulation results demonstrate that the proposed algorithm can achieve better performance than other baseline algorithms, and keep CPU temperature under control.