在合成孔径雷达地面动目标检测系统中,需要无模糊地估计运动目标的径向速度。当径向速度接近最大不模糊速度时,传统多通道最大似然方法的估计性能下降。针对以上问题,提出一种基于距离频率干涉相位的最大似然估计方法。相较于传统方法,所提方法有效地降低了概率密度函数扩散的影响,当径向速度接近最大不模糊速度时能够实现准确估计。从相关系数和干涉相位两个方面分析了噪声对径向速度估计的影响机理,并通过增加独立样本来减小噪声的影响。
The radial velocity of a moving target should be estimated unambiguously in the synthetic aperture radar-ground moving target detection system. Estimation performance of the conventional maximum likelihood method based on multi-channel would decline when the radial velocity approaches the maximum unambiguous velocity.To solve these problems,the ML estimation method based on the range frequency interferometric phase is proposed in this paper.Compared with the conventional method,the proposed method weakens the effect of probability density function spreading.The radial velocity can be estimated accurately even when it approaches the maximum unambiguous velocity.The effect of noise on radial velocity estimation is also analyzed from two aspects of correlation coefficient and interferometric phase,which can be reduced by increasing independent samples.Numerical and real data are processed to demonstrate the effectiveness of the proposed method.