The paper discussed the method of soil moisture inversion using the ALOS/PALSAR data whose study region was Daxing area in Beijing. For ALOS/PALSAR data, co-polarization backscattering coefficient was calculated using the AIEM model, and the Oh model was used to describe the characteristics of cross-polarization scattering. According to backscatting data, the difference between co-polarization backscattering coefficient and cross-polarization backscattering coefficient didn’t change with the soil moisture, so we can build the functional relations between backscattering coefficient and roughness parameters. The bare soil moisture was retrieved using BP neutral network, and the simulated data was used to validate the accuracy of the model. The result showed that the estimated soil moisture error was 0.035 m3/m3, and the relative error was 13.9%. So L-band radar data can be used to inverse soil moisture in the bare region. The research result could provide information for agricultural irrigation, disaster monitoring and environmental assessment.