LoRa low-power wide-area network protocol has recently gained attention for deploying ad-hoc search and rescue (SaR) systems. They could be empowered by exploiting body-UAV links that enable communications between a body-worn radio and a UAV-mounted one. However, to employ UAVs effectively, knowledge of the signal's propagation in the environment is required. Otherwise, communications and localization could be hindered. The radio range, the packet delivery ratio (PDR), and the large- and small-scale fading of body-UAV LoRa links at 868 MHz when the radio wearer is in a Mediterranean forest are here characterized for the first time with a near-ground UAV having a maximum flying height of 30 m. A log-distance model accounting for the body shadowing and the wearer's movements is derived. Over the full LoRa radio range of about 600 m, the new model predicts the path loss (PL) better than the state-of-the-art ones, with a reduction of the median error even by 10 dB. The observed small-scale fading is severe and follows a Nakagami-m distribution. Extensions of the model for similar scenarios can be drawn through appropriate corrective factors.

翻译：LoRa低功耗广域网协议近年在构建自组织搜救（SaR）系统中备受关注。通过开发人体-无人机链路，可实现在人体佩戴无线电装置与无人机载无线电装置之间的通信，从而增强系统性能。然而，要有效部署无人机，必须掌握信号在环境中的传播特性，否则通信与定位功能将受到制约。本研究首次针对佩戴者在森林环境中的情况，对868 MHz频段人体-无人机LoRa链路的无线电覆盖范围、数据包投递率（PDR）以及大尺度与小尺度衰落特性进行了表征，其中无人机采用近地飞行模式，最大飞行高度为30 m。推导出考虑人体遮蔽效应与佩戴者运动的对数距离模型。在约600 m的全LoRa无线电覆盖范围内，新模型对路径损耗（PL）的预测精度优于现有最优模型，中位误差降幅达10 dB。观测到的小尺度衰落具有严重性，且服从Nakagami-m分布。通过引入适当的修正因子，该模型可拓展至类似场景。