基于压缩射频层析成像实现的人体运动行为智能感知理论与方法研究

This project mainly concerns the theories and methods on intelligent motion acquisition with compressive radio tomographic imaging, which facilitates a new sensing paradigm for easy wireless sensor network implementation, compressed motion sensing capability and supportive multi-level motion analysis. In particular, the sensing model of radio tomographic imaging and parameter estimation will be studied. The compressed sensing based optimization strategy of sensing efficiency will be explored for radio tomography imaging. The multi-layer radio tomography imaging network and algorithm for multidimensional image reconstruction will be established for multi-level motion sensing. The efficiency and multiple-cuing acquisition of radio tomography imaging system are the most key issues in this project, which can be fully addressed via compressive sensing based heuristic link-scanning technology and multi-layer radio tomography imaging network. While the contributions of this project will be a novel yet practical solution for intelligent surveillance, it will also substantially contribute to the development of the radio tomographic imaging. The studies will have definitively significances in various applications involving public surveillance, healthcare monitoring, anti-terrorism and rescue so on.

研究基于压缩射频层析成像实现的人体运动行为智能感知理论与方法，探索适合分布式无线传感器网络实现方式、具备压缩运动信息集成获取能力、支持多层次运动行为分析综合的新智能感知模式。特别地，研究射频层析成像的传感模型和模型参数估计方法；探究压缩传感支配的射频层析成像传感效率优化策略；探索满足多层次运动行为感知的压缩射频成像采样模型，及有效的多维状态影像重构算法。射频层析成像实现运动行为感知的传感效率和多线索获取，是本项目的核心问题，解决问题的关键在于建立压缩传感支配的启发式射频链路扫描技术和满足多层次行为信息获取要求的多维射频层析成像网络采样模型。此方面的研究，即是填补了智能监控对多层次运动行为信息获取的迫切需要，也是射频层析成像技术发展的必然要求，具有重要的学术意义，在公共监控，医疗护理监控，反恐求援等领域具有广泛的应用前景。

本项目以实现泛在的人体运动行为智能感知为目标，以无线传感器网络实现的窄带射频链路采样为传感手段，分别从射频层析采样成像的传感模型和模型参数估计方法，压缩传感支配的射频层析成像传感效率优化策略，及满足多层次运动行为感知的射频采样模型和多维特征影像重构算法三个方面，重点研究了限于人体存在状态及运动特征的室内和室外无源定位技术，人体日常行为识别及异常行为监测技术，小型人群下的个体生物特征获取和识别等，突破了传统传感方式感知效率、通信带宽及计算资源受限的瓶颈，形成了灵活分布式网络化实现、具备压缩运动特征信息集成获取能力、支持多层次运动行为分析综合的新智能感知模式。..此方面的研究，无论是对发掘和利用已有泛在射频链路从数据传输到测量工具功能性复用的延拓，还是对射频层析采样成像技术的应用创新，都具有重要的学术意义，在公共监控及安全，医疗护理监控，反恐救援等领域具有广泛的应用前景。