基于扩展卡尔曼滤波的机动目标跟踪研究

针对雷达导引头跟踪机动目标问题,建立了一种适于高机动目标探测的非线性系统模型,采用扩展卡尔曼滤波算法推导适用于弹载脉冲多普勒雷达机动目标探测的滤波模型,用以估计俯仰失调角、方位失调角、弹目相对速度和距离等信息,利用系统数学模型对目标加速度及弹目视线角速率进行解算估计。对标称弹道和极限弹道的跟踪仿真表明,所建立的系统模型和滤波设计方案,能够处理真实飞行中存在的部分信息缺失问题,对于极端情况下的遮挡问题也能在一定时间内保证较高的跟踪精度。     

深空探测火箭的发展及应用

阐述了国外深空探测火箭的种类、结构及技术性能，对其、发展现状及趋势进行了论述，并简要介绍了国内探测火箭的发展状况。     

遥测系统的多普勒频偏盲估计

提出一种适应于在加性高斯白噪声信道(AWGN)下遥测系统的多普勒频偏盲估计算法。针对不同调制方式和波特率的遥测信号在较大频偏范围内对多普勒频偏进行估计,对算法具体实现时参数的选择及影响频偏估计精度的因素进行分析。     

基于频域的软件接收机多普勒补偿算法

由于卫星和接收机之间存在的相对运动,在接收机使用过程中必然会引入多普勒频移。而多普勒频移将导致扩频增益发生衰减,影响捕获工作的完成,因此在捕获过程必须进行相应的补偿处理,去除多普勒频移所产生的影响。本文以时域复指数相乘等于频域偏移原理为基础,采用频域补偿算法替代时域串行搜索算法改进捕获算法,简化了计算过程,节省了频域变换处理计算量,通过实际算例证明有效可行。     

基于深度学习的毫米波和亚毫米波成像仪的图像增强技术

风云四号卫星毫米波和亚毫米波成像仪(MMSI)数据根据采样方式分为过采样和非过采样数据。由于采样方式的影响,非过采样数据在采样过程中会有一定的信息损失。为解决采用简单的线性插值方法做精细化处理时提升精度有限问题,采用基于深度学习的方法增强MMSI亮温图像,设计卷积神经网络重建风云四号卫星MMSI的亮温图像和风云三号卫星微波成像仪亮温图像。实验结果显示:相比传统的双三次插值方法,在风云三号卫星微波成像仪亮温图像样本上峰值信噪比提升了1.13dB,结构相似度提升了0.01。实验结果表明:对于非过采样亮温数据,采用基于深度学习的方法增强图像具有更高的精度,同时可在其他微波探测仪数据中使用,具有很强的普适性。     

星载相控阵GNSS-R测高系统设计与实验

星载GNSS-R测高系统采用多波束相控阵天线和干涉互相关处理方式,接收北斗导航卫星和GPS导航卫星发射的L波段导航直射信号与海面反射信号,进行海面高度探测。详细介绍GNSS-R测高系统设计与实现、集成和外场试验等,外场试验不仅验证了相控阵GNSS-R测高系统功能,更重要的是验证了核心软件算法的正确性。GNSS-R测高系统适用于中尺度海洋现象的观测,可在时间和空间尺度上扩展雷达高度计的观测能力,并与之相互补充,有利于对复杂海面的中尺度结构进行较高时间分辨率的观测。GNSS-R测高仪为星载高精度海面高度探测提供了新型载荷。     

运动的核弹比静止的核弹威力小?——多普勒效应、红移、光行差和动体的质能当量关系

证明爱因斯坦立足于光速不变假设和长度收缩假设的洛伦茨变换未能正确地解释多普勒效应、红移和光行差现象;他也未能给出正确的动体的质能当量关系.作者立足于完全符合相对性原理的伽利略变换,成功地解决了上述问题.     

Beam-Park-Experiment-1/2000 with TIRA

A monostatic 24-h debris observation campaign (BPE-1/2000) has been prepared and conducted using FGAN's TIRA L-Band system. Based on experiences from previous Beam-park experiments a similar largely automated data processing is applied on an extended range window of 300–2000km. More than 1500 detections are encountered, 471 of them are verified as being real objects in Low-Earth-Orbit (LEO). PROOF's observation forecasting of catalogued objects is evaluated against the observed objects, and the difficulties obtaining radar cross-sections (RCSs) and object sizes from Beam-park experiments are discussed. Sidelobe detections are identified by using background information like two-line element (TLE) sets and/or catalogued RCSs.In comparison with previous experiments, the statistics show similarities confirming the concept of Beam-park experiments for space debris observations, despite the snapshot character of 24-h experiments. The comparison with MASTER/PROOF'99 and ORDEM2000 leads to a reasonable agreement between models and observations.     

Coronal radio-sounding detection of a CME during the 1997 Galileo solar conjunction

Frequency fluctuations of the Galileo S-band radio signal were recorded nearly continuously during the spacecraft’s solar conjunction from December 1996 to February 1997. A strong propagating disturbance, most probably associated with a coronal mass ejection (CME), was detected on 7 February when the radio ray path proximate point was on the west solar limb at about 54 solar radii from the Sun. The CME passage through the line of sight is characterized by a significant increase in the fluctuation intensity of the recorded frequency and by an increase in the plasma speed from about 234 km s⁻¹ up to about 755 km s⁻¹. These velocity estimates are obtained from a correlation analysis of frequency fluctuations recorded simultaneously at two widely-separated ground stations. The density turbulence power spectrum is found to steepen behind the CME front. The Galileo radio-sounding data are compared with SOHO/LASCO observations of the CME in the corona and with WIND spacecraft data near the Earth’s orbit.     

一种基于稀疏分解的微多普勒频率估计算法

目标微动特性和微多普勒特征分析在真假目标识别方面发挥了重要作用。由于微动目标雷达同波具有非线性、多分量性等特征,需要相应的具有高分辨力、低交叉项、大的动态范围的分析工具,才能较好地揭示目标微多普勒特征。稀疏分解方法中的匹配追踪（MP）具有频域高分辨能力,对于信号细微特征提取具有很好的效果。研究了基于匹配追踪的微多普勒频率估计问题,该方法可以准确提取出目标微多普勒频率,为后续的目标识别提供了重要的依据。-