Optical interferometers are most widely used in precise displacement measurements, such as semiconductor metrology and lithographic applications, for their capability of obtaining sub-micrometer resolution and non-contact measurements. In optical interferometry, the measured physical quantities are encoded in the phase of interference or fringe patterns. Many techniques, extracting the phase information from the interference patterns, have been proposed to improve the performance and accuracy of interferometers measuring linear displacement. However, these techniques inherently include the nonlinearity and systematic errors of optical interferometers. Hence, to measure the displacement with nanometric resolution, these error sources have to be identified and corrected. We propose a new signal processing algorithmm, extracting the phase information from the spatial interference pattern, for precise displacement measurement. The configuration of the proposed system is similar to Twyman-Green interferometers except that the reference mirror is slightly tilted to generate equally-spaced fringes. This method is to measure the fringe movement by detecting the fringe peak positions which are linearly proportional to the displacement of an object. The system acquires an image of interference fringes on the CCD to find fringe peak positions and traces the fringe movement to obtain the displacement of the target mirror. The peak position is determined by fitting the intensity signal to the quadratic curve around the peak positions. This study shows a simple method capable of detecting the fringe movement from the fringe peak positions. Also, we present the analysis of error occurred in the proposed method and propose the compensation method for the major error sources.
