The Research On Interference Phase Unwrapping Algorithms For Precise Measurement Of MEMS Device

Phase unwrapping algorithm was originally applied for INSAR initial data analysis, and then it was applied for the topographic map drawing. Since the mid-1990s, with the development of MEMS technology,phase unwrapping algorithms gradually developed. Now it has become an essential element for stroboscopic microscopic interferometry system (SMIS)to measure MEMS surface accurately and it is greatly related to parameter setting and testing accuracy of the equipment. This project mainly does research on phase- unwrapping algorithms for recovering surface topography of MEMS-DMs precisely.Firstly the development of phase unwrapping algorithms,resource and background of the project and the main content of thesis are presented, and then path tracking method and minimum norm method are presented on the basis of the principle of the phase unwrapping.This project is focused on least-squares phase unwrapping algorithms. At first the principle of least-squares phase unwrapping algorithms is introduced; secondly the ideas of discrete cosine transform and multi-grid algorithm and their respective algorithm steps are expounded. Then discrete cosine transform and multi-grid algorithm are separately applied for phase unwrapping least-squares algorithm for recovering surface topography; thirdly convergence rates and standard deviation values of these two algorithms are compared, then weighted least squares phase unwrapping by means of multi-grid algorithm techniques is presented with great advantage for phase data with large shear discontinuities.Finally the experimental result shows that weighted least squares method used in phase unwrapping can be used to recover MEMS-DMs surface topography pretty well and it does a great favor for the following testing of characteristics of MEMS-DMs.