Abstract: Soil moisture (Ms) is an important parameter for agricultural, meteorological and hydrographic studies. In order to monitoring soil moisture better, this paper is focused on the Ms estimation methodology based on the multi-polarization and multi-angle advanced synthetic aperture radar (ASAR) data. For the bare soil surface, the inversion model was built based on the simulated data using Advance Integral Equation Model (AIEM). In a wheat-covered area, the HH polarization backscattering coefficient at a low incidence angle was significantly positively correlated with Ms, however, the HH polarization backscattering coefficient at a high incidence angle was significantly positively correlated with the vegetation water content (Mv). Mv is an important parameter of the water-cloud model, so the ASAR data at HH polarization and high incidence angle was used to separate the contribution of wheat canopy backscatter coefficient from the total SAR backscatter coefficient. Then a semi-empirical model was developed to estimate Ms in a wheat-covered area based on the multi-angle ASAR data. Validation was performed by the measured data, the precision of Ms estimation for the bare soil surface and the wheat-covered area was 3.55% and 3.81% respectively. The semi-empirical model has a quite good estimation precision.