Abstract：Global Navigation Satellite System Interference and Reflectometry (GNSS-IR) usually uses single-satellite single-frequency reflection signals as data sources for soil moisture inversion,but its inversion accuracy is limited and it is easy to ignore the differences and complementarity between different satellites dual frequency reflection signals.In order to further improve the retrieval accuracy of soil moisture in the mining area,taking Dongchen Ecological Park in Panji Mining Area,Huainan City,Anhui Province as the research area,GPS satellite signals and soil moisture data were collected through HD-V8 receiver and soil moisture monitoring system (HZR80 type).The GNSS-IR soil moisture retrieval method was used to obtain the interference characteristic parameters (amplitude,frequency,and phase) of the reflection signals of the L1 and L2 bands of the GPS PRN1 and PRN22 satellites,and the correlation analysis was performed with the measured soil moisture values.According to the correlation analysis results,the interference characteristic parameters with strong correlation are selected as the optimal inversion parameters.The adaptive weighting algorithm is used to determine the optimal weighting factor for data fusion of the optimal inversion parameters; and then the optimal inversion parameters of the L1 and L2 bands of the PRN1 and PRN22 satellites are used.Three soil moisture inversion models based on the single frequency method,the mean method and the fusion method was established using the optimal inversion parameters of L1,L2 bands,the arithmetic mean of the optimal inversion parameters,and the fusion values and confirmed the model accuracy.The results show that the correlation between the amplitude and the measured value of soil moisture is stronger,and the absolute value of the correlation coefficient is between 0.556 8 to 0.748 3.It is reasonable to choose the amplitude as the optimal inversion parameter.Compared with the single-frequency method and the mean method,the fusion method has higher accuracy of soil moisture inversion model.Its R2 is 0.763 8,the model verification R2 is 0.936 9,the RMSE is 1.907 8%,the Mean Absolute Error is 1.380 6%,and Maximum Relative Error is 18.848 28%,which indicates that the GNSS-R soil moisture retrieval method based on dual-frequency data fusion can improve the retrieval accuracy.
徐良骥，刘悦，谌芳，张坤. 基于GNSS-R技术的矿区复垦地土壤湿度反演方法研究[J]. 煤炭科学技术, 2020, 48(4): 129-135.
XU Liangji，LIU Yue，CHEN Fang，ZHANG Kun. Study on soil moisture inversion method of reclamation land in mining area based on GNSS-R technology[J]. Coal Science and Technology, 2020, 48(4): 129-135.