White Light Interference Signal Processing Method Based on Fourier Frequency Spectrum Analysis

Sampling noise affects each frequency of white light interference signal Fourier frequency spectrum. Selecting an appropriate filter method will help improve the signal-to-noise ratio of a white light interference signal and obtain highly accurate solution results. In this study, an energy concentration constraint under a certain energy ratio is constructed, and a cost function for selecting the bandpass filter is established. The part calculation process and results of the cost equation under white light LED illumination are given in the experiment. When the cost function reaches its minimum value, the standard deviation of surface roughness is only 0. 02 nm. Finally, a silver mirror and a step structure are measured under the selected parameters. The results show that the proposed method has good performances with respect to repeatability and resolution.