基于MUSIC算法的Pol-InSAR相位估计方法

Estimation Method for Pol-InSAR Multi-Interferometric Phase Based on the MUSIC Method

提出了一种基于MUSIC算法的Pol-InSAR多干涉相位估计方法。该方法结合维纳滤波理论构造最佳加权矢量,通过对其组成的最佳协方差矩阵特征分解得到信号子空间和噪声子空间,充分利用相应像素对其相邻像素的相干信息,并根据MUSIC算法中信号子空间和噪声子空间的正交性构造空间谱函数,通过谱峰搜索准确地估计Pol-InSAR多极化通道相应像素间的干涉相位。该方法具有自适应图像配准和降低相位噪声功能,实测数据的处理结果证明了该方法的有效性。

In this paper we propose a method to estimate the Pol-InSAR multi-interferometric phase based on the MUSIC method. In the method we construct the optimum weighted vector with the Wiener filter which takes advantage of the coherence of neighboring pixel pairs. The method employs the orthogonality of the joint subspace onto the corresponding joint noise subspace by Eigen-decomposition to construct the space spectral function. Therefore, it can accurately estimate the interferometric phase in the corresponding pixels of the multi-polarimetric channels. And it can adaptively coregister the SAR images. The effectiveness of the method is tested using real data.