基于全极化通道融合的目标进动周期提取方法

A precession period extraction method for space target based on full polarization channel fusion processing

摘要:

采用微波暗室静态测量数据动态反演，获取了两类模拟弹道目标在典型战情下的全极化雷达散射截面(RCS)回波序列。针对该时间序列具有的典型非平稳性、拟周期性特征，直接利用谱分析法提取进动周期效果较差等问题，本文在预处理环节利用小波方法进行降噪和加强进动通带细节；针对仅利用单一通道进行特征提取所面临的提取精确度不高、极化信息利用不充分等问题，分别利用主成分分析法和极化不变量法进行全极化通道融合处理。实验结果表明，利用极化信息提取方法明显优于单极化通道提取方法，在无噪观测环境下误差小于0.002 Hz。

Abstract:

The fully polarized Radar Cross Section(RCS) sequences of two simulated ballistic targets are obtained from microwave anechoic static measurement inversion under typical combat conditions. In view of the nonsmoothness and proposed periodicity characteristics of this time series, it is not effective to estimate precession period directly by spectral analysis. Therefore, wavelet method is employed to reduce noise and enhance the details of the precession passband in the pre-processing process. The full polarization channel fusion process is carried out by Principal Component Analysis(PCA) and polarization invariant method respectively. The experiment results show that the extraction method using polarization information is significantly better than the single polarization channel extraction method, and the error is less than 0.002 Hz in noiseless observation environment.