A three-dimensional (3D) scattered data filtering method in combination with averaging process is proposed to improve the signal-to-noise (S/N) ratio of space charge raw signal under periodic electric field. The spatial relationship between each point and its neighborhood is constructed based on the 3D information of space charge, i.e., phase, amplitude and position. Two 3D filtering methods, i.e., Gaussian kernel convolution method and moving least squares surface projection method, are used to filter the feature and non-feature areas of the space charge distribution, respectively. The effectiveness of 3D filtering is verified via analyzing the responses of space charge signal in PMMA under various periodic electric fields. In addition, the results indicate that the number of averaging times can be appropriately adjusted based on the intensity of the polarization electric field.

中文翻译：

---

周期性电场下电介质空间电荷信号处理的3D滤波研究


提出了一种结合平均处理的三维（3D）散射数据滤波方法，以提高周期性电场下空间电荷原始信号的信噪比。基于空间电荷的3D信息，即相位、幅度和位置，构建每个点与其邻域之间的空间关系。采用高斯核卷积法和移动最小二乘表面投影法两种3D滤波方法分别对空间电荷分布的特征区域和非特征区域进行滤波。通过分析PMMA中空间电荷信号在不同周期电场下的响应，验证了3D滤波的有效性。此外，结果表明，可以根据极化电场的强度适当地调整平均次数。