Inversion of rough surface parameters based on Support Vector Machine

This paper emphasis on the inversion of rough surface(sea surface) parameters based on Support Vector Machine (SVM) method, mainly including inversion of wind speed above the rough surface(sea surface), the root mean square(RMS) height and the correlation length of the rough surface. The far-field backscattering of the rough surface is calculated utilizing the method of MOM for horizontal and vertical polarizations. Then the training samples are established with the scattering coefficient as the characteristic value and the parameter to be inversed as the target value, and the relevant parameters of the rough surface (sea surface) are inversed successfully for two polarizations. The impact which results from training samples on the inversion is discussed simply.