中文摘要：

SAR地面慢动目标检测是利用SAR实现空间对地观测应用的一个主要方面,具有重要的理论意义和迫切的实际需求.目前,通过对协方差矩阵进行特征值分解得到三个检测量（第二特征值、干涉相位和相似度）,为慢动目标检测的实现开辟了一条新途径.本文针对相似度和第二特征值检测量存在的理论缺陷,首先,根据对角化矩阵与Pauli自旋矩阵构成的单位上半球面点具有一一对应的特点,从理论上修正了相似度检测量表达式.进而,首次提出并证明了＂对协方差矩阵进行邻域平均预处理是第二特征值作为有效动目标检测量的前提条件＂这一命题,有力地补充和完善了特征值分解检测量的理论体系.仿真实验结果也证明了理论推导的正确性.

英文摘要：

In air-to-ground synthetic aperture radar（SAR） surveillance,it is desirable to be able to detect slow moving targets within strong ground clutter.At present,using eigen-decomposition to the sample covariance matrix,three detection metrics（second eigenvalue,interferometric phase,similarity） can be acquired,which provides a new approach of detecting the slow ground moving targets.The paper aims at the theoretical shortages of the similarity and second eigenvalue.Firstly,according to diagonalizing matrix corresponding to a point on the upper-half of the unit sphere if the Pauli spin matrices ale chosen as basis vectors,the paper revises the expression of similarity detection metric. Secondly, the paper originally puts forward and proves a proposition, which is the second eigenvalue＇ s validity tightly associated to the adjacent average processing to the sample covariance malrix. The two conclusions effectively reinforce the theoretical system of eigen-decomposition detection metrics. The simulated results prove the correctness of the conclusions.