基于时间反转矩阵理论的障碍物形态数据的确定

 针对经典多重信号分类法（MUSIC）算法存在图像重建包含目标信息较少的问题，引入时间反转矩阵推导建立确定散射物图像的理论方法，分析改进MUSIC 算法的定位精确性及其与阵元间距信噪比信号入射角度差间的关系。应用Ipswich 数据集，试验得出改进算法实现了较好的数据图像重建，为算法在确定障碍物形状上的优势提供依据。通过对非对称的双圆形金属柱体对比试验，得出算法在高频情况下可以较好的完成图像重建，且频率越高图像重建的特征值越小，分辨率越高。

Determination of Morphological Data of Obstacles Based on Time Reversal Matrix Theory

Image reconstruction using classic multiple signal classification (MUSIC) algorithms contains little target information. To solve this problem, time reversal matrix theory was adopted to obtain the image of scattering objects. The positioning accuracy of the modified MUSIC algorithm and its relationship with array element space, signal-to-noise ratio, and difference of signal incident angle were analyzed. Ipswich experimental datasets indicate that the improved image reconstruction algorithm realized better image reconstruction, providing evidence for advantages of the algorithm and other algorithms in determination of the morphology of obstacles. Comparative experiment on non-symmetric double-circle metal cylinder shows that the algorithm achieved better image reconstruction in high frequency conditions, and the higher the frequency, the smaller the eigenvalue of image reconstruction, and the higher the resolution.