近年來，基於無線電的人體感測引起了大量的研究關注，並具有廣泛的應用，例如電子醫療監控、室內安全和工業監控。大多數現有研究都集中在分析固定接收器收集的無線信號擾動上。在本論文中，我們展示了 UH-Sense，這是第一個使用無人機進行人體偵測和定位的系統，其中安裝在無人機上的全向性天線被用於測量來自周圍 WiFi 基地台的信號強度。為了克服無人機引起的噪聲，我們提出了一種新的基於數據驅動之學習方法，用於在無法得知噪聲特徵的情況下對損壞數據進行去噪。接著，我們為無人機開發了基於無線電斷層成像的定位模型，無需收集指紋數據庫即可定位周圍的人。我們顯示 UH-Sense 可輕鬆實現於現代商用平台上，並評估其在不同實際環境中的性能，包括不規則基地台之部署和非可直視之場景。實驗結果顯示，UH-Sense 實現了高偵測性能，其人體偵測性能的平均 F1 分數為93\%，並且具有與使用固定接收器收集之乾淨數據相似甚至更好的定位性能，這是目前既有的去噪方法都無法實現的。;Radio-based human sensing has attracted substantial research attention and enjoyed wide applications such as e-healthcare monitoring, indoor security, and industrial surveillance. Most of the existing studies focus on capturing the perturbations of the wireless signals collected at a fixed receiver. In this work, we present UH-Sense, the first unmanned aerial vehicle (UAV) based human detection and localization system in which an omnidirectional antenna mounted on the UAV is used to measure the signal strength from the surrounding WiFi access points (APs). To overcome the multi-source UAV-induced noises, we propose a novel data-driven learning-based approach to denoise corrupted data without knowing the noise characteristics. Then, a localization model based on radio tomography imaging (RTI) is developed for the UAV to localize surrounding human without collecting the fingerprint database. We demonstrate that UH-Sense is readily deployable on commodity platforms and evaluate its performance in different real-world environments including irregular AP deployment and non-line-of-sight (NLOS) scenarios. Experimental results suggest that UH-Sense achieves a high detection performance with an average F1 score of 0.93 and yields similar or even better localization performance than that of using clean data (i.e., data collected at a fixed receiver), which has not been achieved by any of the state-of-the-art denoising methods.
