To successfully use reflectivity data from X-band radars for quantitative precipitation estimation and storm-scale data assimilation, the effect of attenuation must be properly accounted for. An alternative approach to dealing with attenuated radar reflectivity which is built into the forward observation operator in an Ensemble Kalman Filter (EnKF) has been developed and tested under CASA (Center for Collaborative Adaptive Sensing of the Atmosphere) X-band radar network. Such approach is first tested using the EnKF Observing System Simulation Experiments (OSSE) framework. A simulated quasi-linear convective system with more widespread precipitation and hence more significant attenuation than in an isolated super-cell is used as the truth. To avoid the problem of giving potentially too much trust on fully attenuated reflectivity, an empirical model for the observation error, which is large for small reflectivity values, is deployed to improve the performance of the approach. Within sets of assimilation experiments, attenuated reflectivity data, radio velocity data are assimilated either separately or together. In OSSEs, results are quite promising. The roles of observation error and model error are also discussed after run a set of experiments which contains model error by using different horizontal resolutions. Then a real case (May 24th 2011 Chickasha tornado) which passed through CASA radar network is chosen to test the effectiveness of this approach. The comparison of using or not such approach will be made and discussed as well.