An energy-efficient multi-stage alternating optimization scheme for UAV-mounted mobile edge computing networks

• Published in: Computing Vol. 106; no. 1; pp. 57 - 80
• Main Authors: Wang, Zhenqian, Rong, Huigui
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.01.2024; Springer Nature B.V
• Subjects: Algorithms; Artificial Intelligence; Cloud computing; Computation offloading; Computer Appl. in Administrative Data Processing; Computer Communication Networks; Computer Science; Convexity; Edge computing; Energy consumption; Information Systems Applications (incl.Internet); Mobile computing; Neural networks; Optimization; Regular Paper; Resource allocation; Software Engineering; Unmanned aerial vehicles; User experience; 65L70; Mobile edge computing; Resource allocation; Task offloading; 35A01; 65L20; Path planning; Unmanned aerial vehicles; 65L12; User equipments; 65L10
• ISSN: 0010-485X; 1436-5057
• DOI: 10.1007/s00607-023-01210-9

As users have higher and higher requirements for the quality of experience, traditional cloud computing is gradually unable to meet the needs of user equipments. Hence mobile edge computing networks mounted by unmanned aerial vehicles are introduced to improve user experience and reduce energy consumption. However, most current work is based on neural networks, which require large amounts of labeled data or long training times. Given these challenges, this paper proposes an energy-efficient multi-stage alternating optimization scheme to reduce the weighted energy consumption of the entire network. We analyze the energy consumption of each device and formulate a non-convex optimization problem. Considering the impact of task offloading, resource allocation, and path planning on network energy consumption, we transform the energy consumption problem into three subproblems. And use the coordinate descent algorithm, interior point method, and successive convex approximation method to optimize them alternately. The simulation results show that the proposed optimization scheme can significantly reduce the network’s total energy consumption.