Aimed at the problems that the parameter setting of traditional DBSCAN algorithm relies on the unreliability of manual experience, and the heterogeneous data clustering effect is poor, A CMDBSCAN model is proposed based on Cloud model. Firstly, according to the characteristics of the distance curve angle mutation, the adaptive neighborhood radius is obtained. And cluster density threshold is set according to the distribution density of radar signal, realizing DBSCAN algorithm run adaptively. And then, in combination with the theory of multidimensional cloud model, the validity of DBSCAN algorithm sorting results is evaluated, and the decision results are used to further optimize parameter setting. The experiment is carried out according to the radar signal received during the complex countermeasure process. The result shows that the algorithm can realize the adaptive sorting of nonuniform radar signals and effectively deal with the “batch” problem in multimode radar signal sorting.