Classifier technology is an important part of radar target recognition application, and its design and experiment play a very important role in radar target recognition. AdaBoost classifier thought is a typical integrated learning method based on Boosting thought. The core idea is to weight the same training set based on the classification recognition results of the previous base classifier, and then to assemble all the base classifiers to form a strong classifier. In this paper, a fusion classifier based on AdaBoost is designed, which combines SVM, KNN and random forest as the base classifier. And through the RCS sequence spatial target classification recognition simulation experiment, the results show that the fusion classifier designed in this paper has the advantage of recognition performance and has good stability compared with the three commonly used classifier technologies. However, fusion classifiers have the disadvantage of being sensitive to outliers and training time, which limits their application in reality to some extent.