基于FRFT的域-包络曲线的雷达信号特征提取及自动分类

提取和补充新的特征参数是解决复杂体制雷达辐射源信号分选和雷达目标识别难题的有效手段，为此该文提出一种基于FRFT的域-包络曲线特征向量的提取方法。该方法通过FRFT搜索得到旋转角域的包络曲线函数，提取出该曲线峰值所对应的值、峰值大小及包络曲线峰度这3个特征，构造新的特征向量，并以此作为经典参数的补充。同时，采用动态聚类法就所提取的特征向量分选空间雷达辐射源信号。大量的仿真结果表明，提取的新特征具有较好的类内聚敛和类间分离能力，还具有较好的抗噪声性能，证实了新特征向量作为信号分选参数的有效性和可行性。

Feature Extraction and Auto-sorting to Envelope Function of Rotation Angle Domain of Radar Signals Based on FRFT

A feature vector extraction method to the envelope function in rotation angledomain of radar signals based on FRFT is proposed in this paper, because extracting the novel radar signal parameters is an effective method to solve the complex radar emitter signals sorting and radar target recognition problem. This method searches the envelope function ofdomain through FRFT firstly, extractsvalue corresponding of the peak, peak value and kurtosis of the envelope function to construct a new feature vector as the complement to classical parameters. At the same time, a dynamic cluster sorting method is used to complete the sorting work to radar emitters. A large number of simulation results show that, the new feature vector not only has strong compactness within clusters and large separation between clusters, but also has good anti-noise performance. And the results verify the feasibility and effectiveness of this new feature vector as the complement to classical parameters.