基于幅相分离的属性散射中心参数估计新方法

A New Method for Parameter Estimation of Attributed Scattering Centers Based on Amplitude-phase Separation

利用属性散射中心(ASC)参数估计来识别目标上的散射结构是实现合成孔径雷达(SAR)自动目标体识别(ATR)的重要步骤。为提高属性散射中心参数估计的速度并抑制杂散影响,该文首先从图像中提取多个属性散射中心,然后分别估计各个属性散射中心的参数。为提高单个属性散射中心的参数估计速率,考虑到其幅度和相位相关参数可分离,该文提出幅度相位分离的属性散射中心参数估计思想,与传统方法相比,该思想使参数估计算法复杂度和参数估计时间降低了1个数量级。引入迭代半阈值(IHT)算法提高参数估计精度。根据各个属性散射中心的参数估计结果可识别目标上各种散射结构并确定其在目标上的位置分布。仿真数据、实测数据以及MSTAR数据集得到的参数估计的高效性和高准确性,验证了该文所提方法的有效性。

Parameter estimation of Attributed Scattering Centers(ASCs) corresponding to scattering geometries on targets plays an important role in Synthetic Aperture Radar(SAR) imaging-assisted Automatic Target Recognition(ATR). To achieve computational savings and clutter suppression, we extract the measurements of several ASCs and estimate the parameters of each ASC separately. To improve the speed of the estimation process, we propose a method for parameter estimation of ASCs based on amplitude–phase separation that considers a reasonable assumption that the amplitude-and phase-related parameters of an ASC can be estimated separately and independently. Through the proposed method, the complexity and time consumed for parameter estimation are reduced by one order of magnitude than the traditional method. The Iterative Half Thresholding(IHT) algorithm is introduced to enhance the accuracy of parameter estimation. The types and locations of scattering geometries on the target are determined using the estimated ASC parameters. Using simulated data, measured data, and MSTAR data sets, the accuracy and efficiency of parameter estimation are improved and the effectiveness of the proposed method is verified.