Abstract: The soil moisture inversion algorithm,which is adopted in the soil moisture and ocean salinity(SMOS) mission,uses the semi-empirical Q/H model to figure out the surface emissivity.The Q/H model shows the effects of the surface roughness on the emissivity at V polarization are same as that of Hpolarization.In this study,we use the advanced integrale-quation model to generate a simulated database with a wide range of the surface roughness and soil moisture conditions under SMOS sensor configurations and develop a simplified multi-angular surface emission model based on the simulated database.Based on the parameterized model,an inversion procedure is set up in terms of dual-polarization microwave brightness temperatures to retrieve soil moisture with the minimum auxiliary information about the ground.The inversion technique is validated with multi-angular ground microwave radiometer experiment data at L-band from several test sites at Beltsville,MD.The accuracy in random-mean-square error is about 4% at inciden angles of 20°～50°.The results reveal that the proposed inversion procedure decreases the perturbing effects of the surface roughness on the soil moisture estimation.