基于多重分数阶傅里叶变换ISAR快速成像算法

A Fast ISAR Imaging Method Based on Mutiple Fractional Fourier Transform

关于机动目标的逆合成孔径雷达成像技术在诸多应用领域扮演着重要的角色,但其应用一直存在一个严重问题——时变多普勒频率,它会在信号回波中引入高阶相位项,如调频率项,如果不对其进行精确补偿,最终获得的ISAR图像质量会出现明显恶化。针对上述问题,大量算法相继被提出用以实现调频率的估计,但是这些算法往往存在运算量大以及传递误差影响等问题。针对这些问题,本文提出了一种基于多重FRFT变换的ISAR快速成像算法。首先将ISAR回波信号用二次调频(QFM)信号模型进行建模,再利用相干积累广义立方相位函数(CIGCPF)对信号中的线性项与三次项进行估计补偿,随后提出了最小二乘分数阶傅里叶变换(LS-FRFT)的方法,通过多次LS-FRFT联合估计即可独立估计出调频项,这一处理能够有效降低运算量并消除传递误差的影响。考虑到测量误差以及低信噪比的影响可能降低误差的估计精度,在此前LS-FRFT的基础上,我们又提出了加权最小二乘分数阶傅里叶变换(WLS-FRFT)方法来进一步改善参数估计的精度与稳定性。最终通过实测数据验证了所提算法的有效性。

The inverse synthetic aperture radar(ISAR)imaging of maneuvering rotating targets plays an important role in many applications,but it always faces a serious challenge:the time-varying Doppler frequency,which will induce high-order phase terms,especially chirp rate term,degrading the quality of ISAR images significantly if it cannot be compensated properly.Many algorithms have been proposed to estimate and compensate the chirp rate,however,these methods always suffer from heavy computation burden and considerable error propagation effect.To tackle this problem,we propose a novel ISAR imaging algorithm based on multiple FRFT.The received ISAR signal is modeled as quadratic frequency modulated(QFM)signal first,and the linear term and cubic term can be removed by using coherently integrated generalized cubic phase function(CIGCPF).Then we propose the joint least square fractional Fourier transform(LS-FRFT)method to estimate the chirp rate term independently,which eliminates the effect of propagation error and reduces the computation load.Considering that measurement error and low signal-to-noise ratio(SNR)may degrade the error estimation accuracy,a weighted least square FRFT(WLS-FRFT)method is developed to improve the estimation accuracy and robustness.At last,real test results are presented to validate the proposed method.