面向无人机集群物联网数据采集的、考虑可靠性与连通性约束的信息年龄优化
Reliability- and Connectivity-Constrained Age-of-Information Optimization for UAV Swarm IoT Data Collection
- Koç University（土耳其科技大学）
- Wireless Networks Laboratory, Department of Electrical and Electronics Engineering, Koç University（土耳其科技大学电气电子工程系无线网络实验室）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对无人机集群物联网数据采集，本文提出拓扑耦合紧急度调度器，联合优化数据新鲜度、通信可靠性与连通性，在标称条件下将服务故障减少约3.5倍，性能优于基线方法。
AI中文摘要：
无人机（UAV）集群为物联网（IoT）数据采集提供了灵活平台，可从分布式传感器集群收集时延敏感测量数据并将其转发至融合中心。此类任务需联合优化数据新鲜度、短包通信可靠性及集群连通性，但现有方法通常优先考虑新鲜度，却忽略其对连通性的影响。本文提出一种拓扑耦合紧急度（TCU）调度器，通过单无人机连通性评分对各集群的信息年龄紧急度加权，在集群间分散采集的同时优先处理过时集群，并通过显式概率约束保障连通性。每个时隙内，所得问题为混合整数非凸规划，本文采用逐次凸近似求解：对二进制分配进行松弛，将有限块长速率、无人机间链路可靠性及连通性评分线性化为一个凸子问题，从而保留其耦合关系，而非在交替块间拆分。整数分配通过匈牙利匹配恢复，并与原始非凸约束进行验证；同时，亏空队列在无需显式长时规划的情况下维持长期覆盖。仿真结果表明，所提调度器在所有基线中始终实现最低平均信息年龄（AoI）和最高服务可靠性，在标称工作条件下将服务故障减少约3.5倍。在连通性与覆盖要求日益严格时，该调度器可维持全服务，而对比方法要么不可行，要么导致集群碎片化，证明了联合优化数据新鲜度、通信可靠性与连通性的有效性。
英文摘要：
Uncrewed aerial vehicle~(UAV) swarms provide a flexible platform for Internet of Things~(IoT) data collection by gathering delay-sensitive measurements from distributed sensor clusters and relaying them to a fusion center~(FC). Between successive updates, the FC estimates the current value of each monitored quantity; however, the estimation accuracy degrades as the last update ages. Such missions require jointly optimizing estimation accuracy, short-packet communication reliability, and swarm connectivity, yet existing approaches typically prioritize freshness while neglecting its impact on connectivity. This paper proposes a topology-coupled urgency~(TCU) scheduler that minimizes the FC's accumulated estimation error while incorporating the swarm's expected algebraic connectivity into the scheduling objective. The resulting connectivity reward encourages idle UAVs to reposition so as to preserve swarm connectivity. For each time slot, the resulting problem is a mixed-integer nonconvex program, which we solve by successive convex approximation. Specifically, the binary assignments are relaxed, while the connectivity reward is represented through a linear matrix inequality on the expected Laplacian. The finite-blocklength rate and the inter-UAV link reliability are linearized in the same convex subproblem, thereby preserving their coupling rather than separating them into alternating blocks. The integer assignments are subsequently recovered by Hungarian matching and verified against the original nonconvex constraints, while a deficit mechanism ensures long-term cluster coverage without requiring explicit long-horizon planning...
