| Summary: | An ultra-wide-band radar module sensor for soil moisture monitoring field experiment is investigated in this paper. For this type of mission critical sensor, it is applied to collect the reflected signals from subsurface of bare soil and sand with different volumetric water contents (VWCs) data (which are calibrated by a time domain reflectometer). This problem is formulated as a mapping from the raw data to the physical parameter. The fuzzy logic algorithm is employed to track the trend of the time series data and after the forecasting becomes stable, the parameters of membership functions in the final iteration are extracted as templates and the VWC values are computed based on a recognition fashion. Two type of fuzzy logic systems (FLSs), namely, type-1 FLS and interval type-2 FLS are employed and compared under the root-mean-square-error. Finally, the accuracy of the soil moisture retrieval is also compared under the mean absolute deviation and the root mean square difference, which demonstrates the effectiveness of the algorithm.
|
|---|
