Fairness-Aware Task Scheduling and Resource Allocation in UAV-Enabled Mobile Edge Computing Networks

Abstract

Unmanned aerial vehicle (UAV)-enabled mobile edge computing (MEC) has recently emerged to provide data processing and caching in the infrastructure-less areas. However, the limited battery capacity of UAV constrains its endurance time, and makes energy efficiency one of the top priorities in implementing UAV-enabled MEC architecture. In this backdrop, we aim to minimize the UAV&#x2019;s energy consumption by jointly optimizing its trajectory and resource allocation, and task decision and bits scheduling of users considering fairness. The problem is formulated as a mix-integer nonlinear programming problem with strongly coupled variants, and further transformed into three more tractable subproblems: 1) trajectory optimization <inline-formula> <tex-math notation="LaTeX">$\mathbf {P_{T}}$ </tex-math></inline-formula>; 2) task decision and bits scheduling <inline-formula> <tex-math notation="LaTeX">$\mathbf {P_{S}}$ </tex-math></inline-formula>; and 3) resource allocation <inline-formula> <tex-math notation="LaTeX">$\mathbf {P_{R}}$ </tex-math></inline-formula>. Then, we propose an iterative algorithm to deal with them in a sequence, and further design a penalty method-based algorithm to reduce computation complexity when the branch-and-bound (B&#x0026;B) algorithm incurs a high complexity to solve <inline-formula> <tex-math notation="LaTeX">$\mathbf {P_{S}}$ </tex-math></inline-formula>. Simulation results demonstrate that our proposed algorithm can efficiently reduce the energy consumption of UAV, and help save 17.7&#x0025; &#x2013; 54.6&#x0025; and 78.9&#x0025; &#x2013; 91.9&#x0025; energy compared with Equal Resource Allocation and Random Resource Allocation. Moreover, it reduces more than 88&#x0025; running time and achieves relatively satisfactory performance compared with B&#x0026;B.