基于零相位谱的雷达目标识别

在基于高分辨雷达距离像的目标识别中，相位信息起着重要作用。文中对用于全景图像对齐的零相位表示法进行了改进，得到一种称为零相位谱的平移不变特征，它保留了除线性相位以外的所有相位信息，并进一步提取其低频部分作为识别特征，基于仿真数据的试验表明，与频谱幅度特征相比其识别性能有了较大提高。

Radar Targets Recognition Based on Zero Phase Spectra

Phase plays an important role in targets recognition based on high resolution range profile. In this paper, zero phase representation is improved to get a new shift-invariant feature called zero phase spectra which preserve all the phase signals but the shift, then its low frequency is used as the recognition feature. The simulation result based on the simulated data shows that the new feature has a much better recognition performance than spectra amplitude.