线性预测数据外推超分辨雷达成像

距离—多普勒成像雷达的距离分辨力和横分辨力分别取决于发射信号的有效带宽和目标相对于雷达视线(RLOS)在相于处理区间内的转角。本文研究线性预测数据外推离散傅里叶变换(LPDEDFT)超分辨成像方法,旨在突破普通的FFT距离多普勒处理的限制,提高距离—多普勒成像雷达的分辨力。LPDEDFT在概念上和计算上都比较简单,分两步进行,先用线性预测方法把观测数据外推到观测窗之外,然后对外推过的数据进行普通的离散傅里叶变换。文中给出了B—52飞机缩比金属模型微波暗室转台实测数据和飞行中的Boeing-727飞机外场实测数据的成像结果。LPDEDFT与普通的傅里叶方法相比,在相同信号带宽和目标总转角的条件下可以得到更高的图像分辨力,或者可以用较小的信号带宽和目标总转角获得相同质量的图像。

Superresolution Radar Imaging with Linear Prediction Data Extrapolation

The range resolution and the cross-range resolution of range-Doppler imaging radars are related to the effective bandwidth of transmitted signal and the angle through which the object rotates the relatively to the radar line of sight (RLOS) during the coherent processing time,respectively. In this paper,the superresolution imaging method of the linear prediction data extrapolation and discrete Fourier transform (LPDEDFT) is investigated for the purpose of surpassing the limitation imposed by the conventional FFT range-Doppler processing and improving the resolution capability of range-Doppler imaging radar. The LPDEDFT supererresolution imaging method,which is conceptually and computationally simple,consists of extrapolating observed data beyond the observation window by means of linear prediction,and then performing the conventional IDFT of the extrapolated data. The live data of a metalized scale model B-52 aircraft mounted on a rotating platform in a microwave anechoic chamber and a flying Boeing-727 aircraft are processed. It is concluded that,compared to the conventional Fourier method,either higher resolution for the same effective bandwidth of transmitted signals and total rotation angle of the object or equal-quality images from smaller bandwidth and total angle may be obtained by LPDEDFT.