Advances in wireless communications are empowering the emerging Internet-of-Things (IoT) applications and services with billions of connected devices. Mobile-edge computing (MEC) has been proposed to reduce the round-trip delay of these applications as IoT devices may have limited computing resources and the resource-rich mobile cloud may be far away. On the other aspect, unmanned aerial vehicles (UAVs) may potentially be employed to improve the quality of service and the channel conditions of users. We thus propose to utilize the UAV as a computing node as well as a relay node to improve the average user latency in the UAV-aided MEC (UAV-MEC) network and formulate the UAV-MEC problem with the objective to minimize the average latency of all UEs. As the UAV-MEC problem is NP-hard, we decompose it into three subproblems. We propose an approximation algorithm with low complexity to solve the first subproblem and then we obtain the optimal solutions of the remaining two subproblems, upon which another proposed approximation algorithm employs these solutions to finally solve the UAV-MEC problem. The evaluation results demonstrate that the proposed algorithm is superior to three baseline algorithms.

中文翻译：

---

无人机辅助移动边缘计算网络中的延迟感知物联网服务配置


无线通信的进步正在为新兴的物联网 (IoT) 应用程序和服务提供数十亿的连接设备。由于物联网设备的计算资源可能有限，而资源丰富的移动云可能距离很远，因此提出了移动边缘计算（MEC）来减少这些应用程序的往返延迟。另一方面，无人机（UAV）可能被用来改善服务质量和用户的信道条件。因此，我们建议利用无人机作为计算节点和中继节点来改善无人机辅助MEC（UAV-MEC）网络中的平均用户延迟，并以最小化平均延迟为目标制定UAV-MEC问题所有UE的。由于 UAV-MEC 问题是 NP 难问题，我们将其分解为三个子问题。我们提出了一种低复杂度的近似算法来解决第一个子问题，然后获得其余两个子问题的最优解，在此基础上提出的另一种近似算法利用这些解来最终解决UAV-MEC问题。评估结果表明，所提出的算法优于三种基线算法。