近年来，小型无人机的“黑飞”、“滥飞”事件频发，严重危害个人隐私、公共场所安全、航空安全及国家安全。基于群智感知思想，本课题提出利用大量用户携带的智能设备获取的声音信号实现快速有效的入侵无人机检测。本课题将综合考虑智能设备的感知能力与精度限制、网络的动态不确定性及用户的隐私保护等多个因素，重点研究检测率最大化的群智感知、多节点数据融合与定位、群智感知和用户隐私保护协同优化等基础理论和关键技术。具体而言，本项目将对无人机声音信号的传播及检测进行建模，利用契约理论、拍卖理论进行用户激励方案设计，探求动态环境中用户位置隐私的收购设计方案，最后搭建出基于群智感知的无人机入侵检测系统平台。本项目的实施，有望为小型无人机入侵检测问题提供一种经济可行的解决方案，具有广阔的应用前景。

Recently, illegal flying of small UAVs has brought a new and serious threat for personal privacy, public, aerial and national security. In this project, a crowdsensing based approach, which makes use of users' intelligent devices for sampling the acoustic signals of small UAVs, is proposed to detect incursive small UAVs quickly and effectively. This approach will take multiple factors into consideration, such as sensing capacity and accuracy limitation of intelligent devices, uncertainty in dynamic network and users’ privacy-preserving. In this project, we will focus on fundamental theory and key technology, such as crowdsensing based on maximizing detection probability, multi-node data fusion and localization, and cooperative optimization for crowdsensing and users’ privacy-preserving. Specifically, the channel and detection scheme of the acoustic signals of small UAVs are modeled. Then, contract theory and auction theory are used to design the scheme of users’ incentive. We also search for an optimal design to buy users’ position privacy in a dynamic environment. Finally, a systematic platform for detecting incursive small UAVs based on crowdsensing is incarnated. The implementation of this project will provide an economical, feasible and promising solution for the detection of intrusive small UAVs.