Abstract: When the radar signal processor is abnormal, the signal of radar is subject to external electromagnetic interference, the radar observation will have strip-shaped interference echoes along the radial direction. Such non-meteorological echoes must be identified and eliminated before quantitative applications such as radar precipitation estimation. In this study, a new algorithm for radial interference echo identification is presented based on fuzzy logic. To adapt to the current volumetric data and the upcoming single radial stream data for operational application, the characteristic parameters only on the radial direction are extracted, and the membership functions are established, so that a value that measures the probability of each bin being affected by radial interference echo is get. The algorithm is evaluated and implemented in operational system, and the results show that it can distinguish radial interference echoes from precipitation echoes very well. Compared with the original operational algorithm, the CSI score is improved by 0.232, which can significantly improve the accurate identification of radial interference echoes and slightly reduce the false elimination of precipitation echoes. It has robust operation ability and can make positive contribution to quantitative applications (such as precipitation estimation).