Based on the original backscattering signal of Micro Rain Radar and the RaProM algorithm, the equivalent radar reflectivity, the particle falling velocity and Doppler spectrum width are calculated after power spectrum calculation, noise removal and deblurring. Furthermore, the precipitation type are identified. Considering particles falling velocity, the equivalent radar reflectivity, particle size characteristics of different precipitation type and whether there is a bright band, RaProM algorithm can identify particle phase including snow, drizzle, rain, hail and mixed type. In addition, the liquid precipitation parameters, such as radar reflectivity factor and rain intensity, are calculated. Subsequently, three typical cases of stratiform cloud precipitation on July 2, 2021, rain-snow conversion on December 25, 2019, and gradually decreasing bright band height on March 4, 2018 are selected to verify and discuss the results. The method of precipitation type classification is applied to typical stratiform precipitation, the vertical structure shows snowflakes in the supercooled water area, mixed type precipitation in the ice-liquid conversion zone near the 0 ℃ layer and liquid precipitation below the bright band, which verified the validity of the method. The methods are further applied in precipitation type classification and bright band detection, and the results show that compared with standard inversion process of Micro Rain Radar, RaProM algorithm has the advantage of no assumption of precipitation type and considering the upward velocity of particles (such as snowflakes). The results of RaProM algorithm are in good agreement with the co-located microwave radiometer and cloud radar in the vertical structure, and the deviations from the ground disdrometer in the raindrop size distribution and rain intensity are reduced compared with the products of Micro Rain Radar.