通过提取和利用雷达脉冲间隔高维特征,提出了一种基于决策树的雷达辐射源识别方法。将相邻脉冲间隔所构成的向量作为脉冲的高维特征,以增强不同雷达信号之间的可分性,再利用聚类方法提取脉冲列中的这种特征;然后将该特征构成特征向量,以表现特征的整体性;随后基于该特征向量构建决策树分类模型;最后将学习到的模型用于未知雷达脉冲列的识别。仿真实验验证了新方法在不同数据量和数据噪声场景下相对于传统方法的显著优势。

In this paper, a radar emitter recognition method based on decision tree is proposed by extracting and utilizing the highdimensional features of radar pulse interval. The vector formed by adjacent pulse interval is taken as the high-dimensional feature of the pulse to enhance the reparability between different radar signals. Such feature is extracted from the pulse column by clustering method, and then the feature is formed into a feature vector to show the integrity of the feature. Then, a decision tree classification model is constructed based on the feature vector. Finally, the model is used to identify the unknown radar pulse train. Simulation results show that the new method has significant advantages over the traditional method in different data volume and data noise scenarios.

TN957. 51