基于GPS信号的土壤湿度反演研究

Soil moisture is the key parameter been concerned in meteorology,pedology,hydrology,ecology and precision agriculture. It has been drawn great concern of our govenment and scientific and technological circles. The traditional monitor approach of soil moisture has some shortcomings such as timeliness sent, low level of automatization, high cost of implementation. The GPS remote sensing is a new concept technology which has been researched early abroad, it has been successfully applyed in oceans and ice layer detection. The research result indicated the optimum electromagnetic wave for soil moisture remote sensing is between 0.1 and 0.3 meter which is the band of GPS signal. This project plane to inversion research the soil moisture based on the GPS signal used for reference the application of GPS for ocean remote sensing. By means of analyse soil reflected feature affecting on GPS signal, incorporation modern digital signal processing approach to get the reflected features of GPS signal. Meanwhile research the relation between soil moisture and normalized power of GPS reflected signal by means of inellectual method such as neural network. Use the relation as a indicator,we can setup the inversion model of soil moisture, and find the influencing factor of inversion model to improve measure precision of the soil moisture based on the GPS reflected signal. Carry out this project should provided theoretical basis of study on soil moisture remote sensing based on the GPS signal.

土壤湿度是气象学、土壤学、水文学、生态学、精准农业等领域都关注的重要变量，已引起我国政府和科技界的高度重视。传统的土壤湿度监测手段存在时效性差、自动化水平低、实施成本高等缺点。基于GPS的遥感是一种全新概念的技术，国外早已开始对其进行研究，并成功应用于海洋和冰层探测。研究结果表明：遥感土壤湿度的最佳电磁波波长在0.1m～0.3m之间，GPS信号波长正好在此范围。本项目拟借鉴基于GPS的海洋遥感原理和方法，进行基于GPS信号的土壤湿度反演研究。通过分析土壤反射特性对GPS信号的影响，结合现代数字信号处理方法获取反射GPS信号的特性；在此基础上利用神经网络等智能算法研究反射GPS信号的归一化功率与土壤湿度之间的关系，以此为指示因子建立土壤湿度反演模型；找出影响反演模型的因素，提高基于GPS反射信号反演土壤湿度的精度。该项目的完成将为实际应用GPS进行土壤湿度遥感研究打下理论基础。

土壤湿度是气象学、土壤学、水文学、生态学、精准农业等领域都关注的重要变量，是地球物理学中需要采集的基本数据之一。GPS遥感是一种被动微波遥感技术，该方法可以充分利用现有的GPS卫星系统，而不需要搭建庞大的遥感发射装置。本课题主要探讨将GPS信号用于土壤湿度探测中的若干科学问题。右旋极化的GPS信号经过发射面反射后转化为左旋极化信号，被天线接收的GPS信号功率与反射面土壤的复介电常数有关，而介质的复介电常数又与土壤含水量有关。不同GPS卫星信号在接收机端的功率表现不一致，该差异包含了不同传输路径下的多种影响因素，建立与反射面特性相关的归一化功率表现形式是进行反演的前提。通过建立二维功率谱相关函数与功率之间的关系式，利用相关峰值建立了功率的归一化模型。该归一化模型具有统计学特征，与局部的大梯度湿度不敏感，对大尺度的土壤湿度有着较好的相关性。通过设计多通道相关延迟映射软件接收机，使用软件算法完成双通道GPS信号的相关提取，并获取功率的相关峰值指标，利用该数据展开了土壤湿度反演的实验研究，以功率相关峰值指示因子建立土壤湿度反演模型；利用GPS的天空视图数据建立入射角对反演模型的关系式，探讨了入射角对反演模型的影响。