可重构智能面辅助的低精度量化大规模MIMO系统的信道估计

针对可重构智能面辅助的低精度量化的大规模多输入多输出(multiple input multiple output, MIMO)系统中信道估计问题进行了研究。该系统的信道估计难点在于可重构智能面由近无源反射天线构成, 没有基带信号处理能力。系统观测值通过低精度的模数转换器量化使信道估计问题变得更富挑战性。本文基于基站-可重构智能面-用户的级联信道推导出等效信道, 并证明在虚拟角域上，该有效信道是结构稀疏信号。提出了基于期望最大化的近邻学习广义近似消息传递算法，从低精度量化的观测值中恢复等效信道。仿真结果表明所提出算法比传统算法具有更好的性能表现。     

基于LabVIEW与SQL server设计的浮标电源远程管理系统

针对现有海洋浮标缺乏对电源有效监控管理的问题，提出了一种基于LabVIEW与SQL server编程，结合C8051F021单片机硬件的方法，设计了岸站服务器的终端监控软件和安装于浮标上的硬件电路。通过网络或北斗通讯等实时通讯方式，实现对浮标上电源的实时监控和管理，提高了浮标电源的利用和管理的能力。     

基于空地协同采样的植被覆盖度随机森林估算方法

基于无人机高光谱影像, 建立地形复杂地区植被覆盖度的非参数随机森林回归估算模型。为获得构建随机森林模型所需的足够数量的训练样本, 利用低空无人机搭载的光学相机, 在从地面难以到达的山地、水域和植被茂密区, 通过垂直拍摄获得厘米分辨率的航拍影像, 作为对地面样方采样的补充。首先计算地面数码相机照片和无人机可见光影像的红绿蓝植被指数(red-green-blue vegetation index, RGBVI), 然后使用大津分割法提取样方的植被覆盖信息, 得到构建模型所需的训练样本。在此基础上, 基于2018年8月16—18日在内蒙古自治区察右中旗油娄沟矿区获取的GaiaSky-mini2无人机高光谱影像数据, 利用递归特征消除算法优选参与随机森林回归的特征变量集, 利用空地协同获取的训练样本构建植被覆盖度的随机森林回归估算模型。该模型在测试集上的确定系数R2为0.923, 均方根误差为0.087, 优于常用的像元二分模型, 可用于矿区植被动态信息的精细化监测。     

相参频率捷变雷达目标稀疏重建性能边界综述

相比频率固定的脉冲多普勒体制, 频率捷变体制在抗干扰方面具有显著优势。但在该体制下, 基于匹配滤波的信号处理算法存在旁瓣平台问题, 难以与动目标检测兼容。压缩感知理论将目标参数估计建模作为欠定方程求解, 为该问题提供了解决思路。在相参频率捷变雷达中, 压缩感知能否准确重建目标是一个基础性问题。本文梳理了针对该问题的相关研究, 借助相变理论与相变曲线的解析表达式, 定量描述了捷变频雷达重建目标的成功概率与主要系统和目标参数之间的关系; 该理论性能边界与仿真实际所能达到的性能相接近。此外, 还探讨了现有成果在实际应用中的价值, 展望了未来研究方向。     

柔性可穿戴传感器及大面积阵列制造新方法

柔性可穿戴传感器作为穿戴式电子系统的重要组成部分，在个人健康监护、人机交互体系以及人造电子皮肤等领域具有广阔的应用前景，是当下最前沿的研究领域之一。文章概要介绍了不同类型的柔性可穿戴传感器的研究现状，简略归纳了柔性传感器中的关键技术参数，同时对柔性传感器大面积阵列制造所面临的挑战及新方法进行了阐述和分析，最后展望了柔性传感器未来的发展趋势。     

Optical sensing of explosives in wastewater by competitive complexation

A facile strategy for optical sensing of explosives with the assistance of 1,3,5-Triazine-2,4,6-triamine(MEL) in aqueous solution is proposed. Upon exposure to MEL, the electron-sufficient amino group in MEL was prone to form complexes with electron-deficient nitro group in explosives. The interaction could be tuned by the amount of MEL to fully consumed nitro-contained explosives dissolved in wastewater, thus prevented the gold nanoparticles(GNPs) from combining with MEL. In comparison with common detection method, the dose-response curve procedure reported here was in good linear relationship from 0 μM to 80 μM with good correlation factor(R~2= 0.9871), and was satisfactorily agreed with the proposed theoretical model. The detection limit(Lo D) was1.1 μM by spectrometer, 3.0 μM visualized by naked eyes and 21.3 nM by calculation. The procedure described in this study could be finished within a few minutes and showed its effectiveness in explosive detection upright on the gold surface in real wastewater, making this proposed method a powerful tool in several sensing platforms.     

微生物废水处理产甲烷的自动控制系统

结合现代的自动控制技术和互联网技术,设计了基于微生物废水处理产甲烷的自动控制系统.选用基于Cortex A9主控芯片的Exynos4412作为控制核心,使用云架构的tomcat服务器进行数据存储,配合现场工业级执行单元与标准专用传感器,实现了微生物培养进而处理工业废水以及提高甲烷生产含量的目标.实际运行结果表明:该系统成功将废水处理至排放标准并使得生产的甲烷含量由52%提升至77%,具有较高的实用价值和应用推广价值.     

Re-situating fieldwork and re-narrating disciplinary history in global mega-geomorphology

In 1985, more than thirty geomorphologists, planetary scientists, and remote sensing specialists gathered at a conference center in Oracle, Arizona, to discuss an emerging area of research that they called “mega-geomorphology.” Building on a conference of the same name held in London in 1981, they argued that new techniques of remote sensing and insights emerging from the study of extraterrestrial planets had created opportunities for geomorphology to broaden its spatial and temporal scope. This new approach was, however, neither unproblematic nor uncontested. In the discussions around mega-geomorphology that took place in the mid-1980s, the perceived conflict between the use of remote-sensing techniques to observe phenomena on vast spatial scales, on one hand, and the disciplinary centrality of fieldwork and field experience to geomorphology, on the other, was a recurrent theme. In response, mega-geomorphologists attempted to re-situate fieldwork and re-narrate disciplinary histories in such a way as to make remote sensing and planetary science not only compatible with geomorphological traditions but also means of revitalizing them. Only partially successful, these attempts reveal that the process of adopting a planetary perspective in geomorphology, as in other earth sciences, was neither straightforward nor inevitable. They also show how the field and fieldwork could remain central to geomorphology while also being extensively revised in light of new technical possibilities and theoretical frameworks.     

基于φ-OTDR光纤振动传感系统模式识别方法

提出一种基于相位敏感光时域反射计(φ-OTDR)的光纤振动传感系统, 提取信号的主波能量、 主瓣时延、 主波峰峰值、 主波Q值、 频谱、 3 dB时延、 主瓣峰值、 中心频率、 二次回波损耗、 波数和半波长度11种特征参量作为参考衡量标准, 用多参量模糊评价矩阵的目标识别方法, 对5组200次实验进行相似目标识别分类. 结果表明,  该方法通过多角度参量进行目标衡量判别, 综合识别率达90.87%, 提高了相似识别的准确性.     

基于图像欧氏距离流形降维的端元提取算法

由于多重反射和散射,高光谱图像中的混合像元实际上是非线性光谱混合传统的端元提取算法是以线性光谱混合模型为基础,因此提取精度不高针对高光谱图像的非线性结构.本文提出了基于图像欧氏距离非线性降维的高光谱遥感图像端元提取方法该方法结合高光谱数据的物理特性,将图像欧氏距离引入局部切空间排列进行非线性降维以更好的去除高光谱数据集中冗余的空间信息和光谱维度信息,然后对降维后的数据利用寻找最大单形体体积的方法提取端元.真实高光谱数据实验表明,提出方法对高光谱图像端元提取具有良好的效果,性能优于线性降维的主成分分析算法和原始的局部切空间排列算法.