Using improved SVM decision tree to classify HRRP

Radar target classification by using high resolution range profile (HRRP) as features has been studied in this paper. Support vector machine (SVM) has been used in range profile classification for its good generalization ability for the pattern classification problem with high feature dimension and small training set. In order to perform multi-class classification, decision-tree-based SVM has been studied. Based on the analysis of the structure and the classification performance of the SVM decision tree, and by introducing the defined separability measure that based on the distribution of the training samples into the formation of the decision tree, an improved algorithm for SVM decision tree has been proposed. The scheme of using the improved algorithm for SVM decision tree to classify HRRP has been studied. Experiments using the simulated range profile datasets prove the effectiveness of our scheme.