一种改进的相位误差计算法

自聚焦主要由两个重要步骤组成，即相位误差估计和相位误差校正，在现有的相位误差估计算法中，秩一相位估计算法是较好的一种，文中针对秩一相位估计算法存在的缺陷提出了一种改进的相位误差估计器，在低仪器时改时算法性能与秩一相位估计算法基本一致，但在信噪比较高时，改进算法能比秩一相位估计算示估计精度高，改进算法的另一优点是比秩一相位估计算法收敛快，若把改进算法代入相位梯度自聚焦算法的算法结构，对窗长选择的稳健性要好一些。

An improved phase error estimation algorithm

Autofocus generally encompasses two distinct steps, i.e. phase error estimation and phase error compensation. Of the existing phase error estimation algorihtms, the rank one phase estimation (ROPE) algorithm is better one. By considering the shortcomings of the ROPE algorithm, an improved phase error estimation algorithm is proposed. In general, the performance of the improved algorithm is consistent with the ROPE algorithm. But within a moderate high signal-noise ratio (SNR) range, the improved algorithm is superior to the ROPE. Antother advantage of the improved algorithm is that it converges more rapidly than the ROPE. If we combine the two algorithms with the phase gradient autofocus algorithm, the improved one is more robust in the selection of the window length.