基于MUSIC算法的GTD模型参数估计

MUSIC-based parametric estimation of GTD model

将能够精确描述高频电磁散射的GTD信号模型引入MUSIC算法,并对MUSIC算法作了相应改进。利用特征分析方法的信号与噪声子空间的正交特性,使改进后的MUSIC算法既能精确估计目标散射中心位置,又能估计散射中心类型,取得了较好的效果。针对各种空间平滑预处理方法对噪声子空间与信号正交特性的影响进行了仿真,指出空间平滑预处理方法会影响噪声子空间及信号子空间结构,进而影响散射中心类型的估计,所以应对散射中心类型的最终结果进行修正。

The GTD model which can accurately describe high frequency electromagnetic back scattering is introduced into the MUSIC algorithm as its signal model, and the MUSIC algorithm is improved in some extent. Utilizing the vertical property between signal and noise subspace, the improved MUSIC algorithm can accurately estimate the type of the scattering centers as well as the position of them, and finally results in good effects. The paper also simulates the impact of the spatial smoothing preprocessing method on the vertical property between signal and noise subspace, and points out that the spatial smoothing preprocessing method can affect the structure of the noise subspace and that of the signal subspace, so the preprocessing method can influence the estimation of the type of the scattering centers, that is, the final result of the scattering centers＇ type should be modified.