首页|基于轨迹聚类的GNSS-IR多系统组合土壤湿度估计方法

基于轨迹聚类的GNSS-IR多系统组合土壤湿度估计方法

扫码查看
全球导航卫星系统干涉反射测量(global navigation satellite system-interferometric reflectometry,GNSS-IR)技术能探测浅层地表的土壤湿度.针对多系统组合的土壤湿度反演问题,基于轨迹聚类方法,从全球定位系统(global positioning system,GPS)、北斗卫星导航系统(BeiDou satellite navigation system,BDS)、格洛纳斯(GLONASS)、伽利略(Galileo)导航系统的信噪比观测数据中提取多径干涉相位,利用经验模型求解轨迹聚类后的土壤湿度估计值,以加权平均方式得到系统组合后的估计结果.结果表明,BDS、Galileo反演精度相当且优于GPS、GLONASS,基于轨迹聚类的多系统组合土壤湿度估计方法的均方根误差为0.0414cm3/cm3,相比于单系统的综合反演精度提升约16.3%,相比于单系统的最佳频段反演精度提升约5.2%,所提方法能有效监测土壤湿度的变化.
A GNSS-IR Multi-system Combination Soil Moisture Estimation Method Based on Track Clustering
Objectives:Global navigation satellite system-interferometric reflectometry can be taken advan-tage of identifying soil moisture of land surface.Methods:Aiming at the problem of multi-GNSS combined soil moisture inversion,we use the phase extracted from the signal-to-noise ratio observation data of global positioning system(GPS),BeiDou satellite navigation system(BDS),GLONASS and Galileo system,and solve soil moisture inversion within consideration of satellite track clustering by using the empirical model.The estimation of multi-GNSS combination is obtained by weighted average method.Results:The inversion accuracy of BDS and Galileo is equivalent and superior to GPS and GLONASS.The root mean square error of the method in this paper is 0.041 4 cm3/cm3,which is about 16.3%and 5.2%lower than that of sin-gle navigation satellite system and the optimal frequency band respectively.Conclusions:The results indi-cate that the multi-GNSS reflection signal estimation method based on track clustering can effectively moni-tor changes in soil moisture.

soil moistureglobal navigation satellite system-interferometric reflectometry(GNSS-IR)signal-to-noise ratio(SNR)multi-system combinationtrack clustering

郑南山、何佳星、丁锐、章恒一

展开 >

中国矿业大学环境与测绘学院,江苏 徐州,221116

自然资源部国土环境与灾害监测重点实验室,江苏 徐州,221116

土壤湿度 GNSS-IR 信噪比 多系统组合 轨迹聚类

国家自然科学基金国家自然科学基金联合重点基金

41974039U22A20569

2024

武汉大学学报(信息科学版)
武汉大学

武汉大学学报(信息科学版)

CSTPCD北大核心
影响因子:1.072
ISSN:1671-8860
年,卷(期):2024.49(1)
  • 11