随着5G网络的部署，基于无线传感网络的免携带设备无线定位（DFL）技术将发挥重要作用。该技术不需要目标携带任何设备即可准确定位，具有成本低，通用性好等诸多优势，已被广泛地应用于入侵检测、老人看护等众多领域，未来将在智慧城市、智能工厂的建设过程中发挥不可替代的作用。然而，现有的DFL定位技术仍然面临诸多挑战，如缺乏对目标的隐私保护机制，以及在多目标、不同体型目标定位方面存在精度低、鲁棒性差等缺陷。为此，本项目展开关于DFL定位技术中的隐私保护与高效定位方面的研究。本项目研究方案包含以下三个步骤：首先，设计适用于无线传感网络的隐私保护方案；其次，构建联邦字典学习模型，结合块稀疏编码和邻近算子技术，开发高效的定位分析算法，并将其应用于隐私保护的DFL定位系统。本项目提出的加强隐私保护的高效DFL定位方案融合了稀疏表示等多种理论与技术，为5G网络场景中的目标定位提供新型运行机制与基础理论支撑。

With the continuous deployment of fifth-generation (5G) communication, device-free localization (DFL) technology based on wireless sensor networks has been proposed and has become a research hotspot. DFL technology does not require the target to carry any device to obtain location information. It has many advantages such as low cost and good versatility, which hereby has been widely used in many fields such as location services, intrusion detection, and elderly care. Such DFL technology will play an irreplaceable role in the future construction of the smart city. In view of the current challenges of DFL technology, i.e., lacking privacy preservation and the poor performance of locating multiple targets and variable body-size targets. This project carries out research towards privacy-preserving-based DFL technology in 5G networks. At first, based on the fingerprint method and sparse representation technology, an effective scheme is proposed to protect network privacy of DFL system. We then propose a federal dictionary learning model to improve the localization performance in locating multiple targets who are with different body sizes. In addition, a block-sparse coding algorithm is designed for improving the performance of multi-target localization. Finally, by combining the aforementioned parts, a high-performance DFL scheme can be developed. The privacy-preserving based DFL scheme proposed in this project integrates multiple theories and technologies, which will enrich and improve the existing DFL technologies. The corresponding research achievements will provide supports for target positioning in terms of a new DFL architecture and the fundamental theory.