基于鲁棒压缩感知的无线网络信道状态信息测量机制研究

Wireless network Channel State Information, including Receiving Signal Strength(RSS) and the emerging CSI metric, are the essential and fundamental for the wireless applications, e.g., wireless networks optimization, indoor localization and device-free sensing. In contrast of their universal application, the channel state information measurement is suffering from tedious and time-consuming collecting process with low data quality, e.g. full of erroneous and noisy values, all of which further compromise the algorithms and applications based on them. Recently, with the booming of device free wireless sensing researches, the channel state information measurement mechanism is further confronting enlarged difficulties and new challenges. To this end, this project plans to take advantage of the recent progress of the robust compressive sensing theory and develop a new channel state information measurement mechanism based on this theory. Specifically, this research proposal is going to develop the fundamental theory for low cost and robust measurement mechanism, an accuracy control mechanism and the specification of this mechanism to wireless network optimization, indoor localization and wireless sensing. This mechanism is envisioning to solve the problems of low efficiency and low data quality for the channel state information measurement process and, in turn, greatly enhance the efficiency and effect of the upper layer applications. The applicant has published several related high quality papers, including INFOCOM and etc., and has also taken part in some related NSFC projects, all of which give good support to conduct this project.

无线网络信道状态信息的测量是众多无线网络服务，如无线网络优化、室内定位及无线感知的基础。测量过程的效率与测量结果的质量直接决定了这些上层应用的实用性。然而该测量过程十分耗时且包含大量的错误及噪音信息，成为制约了上层算法的瓶颈。近年来无线感知的研究与发展对测量的效率及质量提出了更高的要求，衍生出新的挑战。鉴于此，本项目拟结合近年来的压缩感知中最新的鲁棒压缩感知理论，研究基于该理论的无线网络信道状态信息测量机制。首先建立高效率可靠无线信道状态信息测量的基本理论，其次设计具有精确度控制的测量机制，最后结合应用进一步探索机制的扩展性。该机制可望有效的解决无线网络状态信息测量过程的低效及测量结果的低质量问题，大大提升上层算法及应用的效率及结果。申请人已发表过多篇包括INFOCOM等在内的相关高质量论文，并参与过多项与本项目相关的国家自然科学基金项目，为本项目的开展提供了良好的支撑。

无线网络信道状态信息的测量是众多无线网络服务，如无线网络优化、室内定位及无线感知的基础。测量过程的效率与测量结果的质量直接决定了这些上层应用的实用性。然而该测量过程十分耗时且包含大量的错误及噪音信息，成为制约了上层算法的瓶颈。近年来无线感知的研究与发展对测量的效率及质量提出了更高的要求，衍生出新的挑战。鉴于此，本项目拟结合近年来的压缩感知中最新的鲁棒压缩感知理论，研究基于该理论的无线网络信道状态信息测量机制。首先建立高效率可靠无线信道状态信息测量的基本理论，其次设计具有精确度控制的测量机制，最后结合应用进一步探索机制的扩展性。该机制有效的解决无线网络状态信息测量过程的低效及测量结果的低质量问题，大大提升上层算法及应用的效率及结果。项目揭示了基于可靠压缩感知的信号状态信息测量机理，并提出基于主动压缩感感知的新方法。将该基于该机制的无线信号地图测量及众包采集方法，提升了测量的效率及减少了测量的开销，并在此基础上提出基于自编码器的地图恢复方法。构建基于高效信道测量的无线多模态定位方法，开发相应定位系统，大大提升了定位精度。建立基于高效信道测量的感知及身份验证机制及相应系统，通过利用WiFi非接触感知的典型场景。有效的拓展的感知系统的适用范围。.该项目已发表论文14篇，其中SCI收录5篇，EI收录14篇，其中包括1篇一区期刊及一篇CCF A类会议论文；申请发明专利7项；培养硕士生4名；主持与本项目相关的国家自然科学基金项目，博士后基金一等资助，国家博士后特别资助等。