论文部分内容阅读
基于雷达目标距离像,研究时变特征提取和核分类器在雷达目标识别中的应用。由于距离像敏感于目标姿态角的变化,单纯的时域或频域方法难以完整刻画目标的散射特性,因此文中采用时频分析方法,首先提取出距离像时频分布的特征参量,再利用主元分析法降低维数,最后采用基于核的非线性分类器进行目标识别。仿真数据和实测数据表明,该方法具有较好的识别效果。
Based on radar target distance image, the application of time-varying feature extraction and kernel classifier in radar target recognition is studied. Because of the change of target attitude angle, the simple time-domain or frequency-domain method is difficult to characterize the scattering characteristics of the target completely. Therefore, the time-frequency analysis method is used to extract the characteristic parameters of distance-time-frequency distribution firstly, Meta-analytic method to reduce the dimension, and finally the use of nuclear-based nonlinear classifier for target recognition. Simulation data and measured data show that this method has a good recognition effect.