Effective improvement of depth resolution and reduction of ripple error in depth-resolved wavenumber-scanning interferometry

Eigenvalue decomposition and least squares algorithm for depth resolution of wavenumber-scanning interferometry

“…The solid lines represent the EDLSA obtained by Eq. (7). With respect to k 2 , the OPD difference dΛ jΛ 23 − Λ 12 j 0.84 mm at x 0.44 mm and y 0.16 mm is much smaller than the depth resolution δ 2 1.51 mm.…”

“…2(e) and 2(h). If the unwrapped phase evaluated by the CNLSA with respect to k 1 was taken as the reference [7], the resulting unwrapped phase errors at the transection line y 1.17 mm, with respect to k 1 and k 2 , would be those shown in Figs. 2(i) and 2(j).…”

“…The Fourier transform (FT) is widely used in optical depthresolved interferometry applications such as wavelengthscanning interferometry (WSI), swept source interferometry, and spectral OCT [1][2][3][4][5][6][7]. The main drawback of these methods is that their profile depth resolution is limited by the width of the sampling window.…”

“…A 200 nm thick film profile was measured. In 2015, Zhang et al proposed the complex number least squares algorithm (CNLSA) for WSI [7]. Its depth resolution is about onefold higher as compared with that of previously used methods.…”

Estimating the number of layers for precise wavelength scanning interferometry

“…To apply it efficiently, two problems should be considered very carefully. The first one is that the depth resolution is relatively low due to the limited wavelength scanning range [5,6] and several methods are available for this problem [7][8][9][10]. The second one is that the precise number of layers is unknown, resulting in it being difficult to precisely extract the useful interference signals.…”