利用一维像序列时域差分估计目标进动频率

Estimation of target precession frequency based on time-domain difference of sequential HRRPs

针对弹道目标识别中目标运动特征参数估计问题,提出一种一维距离像(HRRP)序列时域差分目标进动频率估计方法,该方法首先对一维距离像进行归一化处理,利用峰值法对齐相邻一维距离像,然后计算相邻一维距离像之间的差分值,获得相邻一维距离像差分值序列,最后采用快速傅里叶变换(FFT)从该差分值序列中估计目标进动频率。该方法具有以下优点:利用相邻一维距离像差分值序列能够增强目标进动频率成份,降低其他频率成份,更有利于进动频率的估计;只需要对齐相邻2个一维距离像,而避免了对齐一维距离像序列所带来的问题。利用由FEKO软件计算的进动目标宽带一维距离像仿真数据进行了实验,结果表明所提方法比常规的加权累积法更有效,估计精度更高。

A novel method for estimation of target precession frequency,namely the time-domain difference using sequential High Resolution Range Profile(HRRP),is proposed for missile target recognition in this paper.First,each HRRP is normalized,and two adjacent HRRPs are aligned by the peak value method.Then,the time-domain difference between two adjacent HRRPs is computed to obtain the difference value series.Finally,estimation of the target precession frequency can be solved by Fast Faurier Transformation(FFT)according to the difference value series.The proposed method has two advantages:The difference method is used to increase the component of precession frequency and reduce the other component of frequency;Only two adjacent HRRPs are needed to be aligned,avoiding the problem caused by alignment of all HRRPs.The experimental results for the simulated HRRPs computed by FEKO software show that the effectiveness and the estimation accuracy of the proposed method are better than ones of the conventional weighted accumulation method.