摘要
全球卫星导航系统搭载的L波段信号对土壤水分特别敏感,非常适合土壤湿度变化监测。以信号功率或是延迟为属性,土壤湿度实际值为标签,结合介电常数经验模型、支持向量机、随机森林算法以及BP神经网络、深度置信神经网络等方法进行模型训练便可以建立基于导航信号的土壤湿度反演模型。随着北斗导航系统的建成,利用其三频载波信号和地球静止轨道卫星(GEO)的独特优势进行土壤湿度反演,可以达到在原有技术基础上进一步提升土壤水分监测效果的目的。为了进一步推广北斗系统在土壤湿度监测方面的应用潜力,本文对全球导航系统反射(GNSS-R)反演土壤湿度的现状和反演土壤湿度的原理进行了介绍和分析,并阐述了利用北斗导航信号作为信号源反演土壤湿度的优点。
The L-band signal carried by the global satellite navigation system is particularly sensitive to soil moisture and very suitable for soil moisture change monitoring.Taking the signal power or delay as the attribute,the actual value of soil moisture as the label,combined with dielectric constant empirical model,support vector machine,random forest algorithm,BP neural network and depth confidence neural network,the soil moisture inversion model based on navigation signal can be established.With the completion of BeiDou navigation system,using its three frequency carrier signal and the unique advantages of geosychronons earth orbit(GEO)can further improve the effect of soil moisture monitoring based on the original technology.In order to further promote the application potential of BeiDou system in soil moisture monitoring,this paper introduces and analyzes the current situation of GNSS-R inversion of soil moisture and the principle of inversion of soil moisture,and expounds the advantages of using BeiDou navigation signal as signal source for soil moisture retrieval.
作者
杨昌智
毛克彪
孙一丹
王一帆
王平
郭中华
YANG Changzhi;MAO Kebiao;SUN Yidan;WANG Yifan;WANG Ping;GUO Zhonghua(School of Physics and Electronic-Engineering,Ningxia University,Yinchuan 750021;Institute of Agricultural Resources and Regional Planning,Chinese Academy of Agricultural Sciences,Beijing 100081;State Key Laboratory of Remote Sensing Science,Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100101)
出处
《高技术通讯》
CAS
2022年第11期1196-1201,共6页
Chinese High Technology Letters
基金
宁夏自治区科技创新团队柔性引进人才项目(2021RXTDLX14)
中央级公益性科研院所基本科研业务费专项资金(1610132020014)
遥感科学国家重点实验室开放基金(OFSLRSS202201)资助项目。
关键词
土壤湿度
机器学习
神经网络
北斗导航系统
三频载波信号
地球静止轨道卫星(GEO)
soil moisture
machine learning
neural network
BeiDou navigation system
three frequency carrier signal
geosychronons earth orbit satellite(GEO)