In this work, we study the UAV-enabled data collection problem for high information freshness in wireless sensor networks, where one UAV is dispatched to collect information of ground Sensor Nodes (SNs). The information freshness is measured by the Age of Information (AoI) of each SN, which is defined as the sum of the SN’s data uploading time and the UAV’s flight time after leaving this SN. Two optimization problems of age-optimal data collection are formulated to minimize the SNs’ maximal AoI and average AoI, respectively. An iterative SN association and trajectory planning policy is proposed to seek the age-optimal solutions via an iterative two-step procedure. Firstly, SN association is performed based on the affinity propagation clustering method with an appropriate weight to find a set of data Collection Points (CPs) at which the UAV hovers to collect data and schedules which SNs to upload in what order. Based on this result, trajectory planning is performed to find the max-AoI-optimal and ave-AoI-optimal trajectories of the UAV along the CPs using dynamic programming or genetic algorithm. With the optimized clustering weight, the proposed scheme can always strike a balance between the SNs’ uploading time and the UAV’s flight time in various scenarios. Simulation results show that the proposed strategy can improve the freshness of information collected from all the SNs.

中文翻译：

---

无人机辅助数据收集，以提高无线传感器网络中的信息新鲜度

在这项工作中，我们研究了启用UAV的数据收集问题，以解决无线传感器网络中信息的新鲜度高的问题，其中派出一架UAV来收集地面传感器节点（SN）的信息。信息新鲜度是通过每个SN的信息年龄（AoI）来衡量的，Ao定义为SN的数据上载时间与离开该SN后的UAV的飞行时间之和。提出了年龄最佳数据收集的两个优化问题，以分别最小化SN的最大AoI和平均AoI。提出了一种迭代的SN关联和轨迹规划策略，以通过迭代的两步过程来寻求年龄最优的解决方案。首先，SN关联是基于具有适当权重的相似性传播聚类方法执行的，以找到一组数据收集点（CP），UAV会在该数据收集点上徘徊以收集数据并计划以什么顺序上载哪些SN。基于此结果，执行轨迹规划，以使用动态规划或遗传算法找到沿CP的无人机的最大AoI最优轨迹和AVE-AoI最优轨迹。通过优化的聚类权重，所提出的方案可以始终在各种情况下在SN的上载时间与UAV的飞行时间之间取得平衡。仿真结果表明，该策略可以提高从所有SN收集信息的新鲜度。使用动态规划或遗传算法执行轨迹规划，以找到沿CP的无人机的最大AoI最优轨迹和AVE-AoI最优轨迹。通过优化的聚类权重，所提出的方案可以始终在各种情况下在SN的上载时间与UAV的飞行时间之间取得平衡。仿真结果表明，该策略可以提高从所有SN收集的信息的新鲜度。使用动态规划或遗传算法执行轨迹规划，以找到沿CP的无人机的最大AoI最优轨迹和AVE-AoI最优轨迹。通过优化的聚类权重，所提出的方案可以始终在各种情况下在SN的上载时间与UAV的飞行时间之间取得平衡。仿真结果表明，该策略可以提高从所有SN收集的信息的新鲜度。