Utilizing statistical-spectral correlations of random metasurfaces for multi-wavelength cryptography

We study correlations between the configuration statistics of random metasurfaces and their spectral response. Our metasurfaces consist of a two-dimensional array of silicon nanopillars with widths sampled from a normal distribution placed on a silica substrate. We explore the effect of tuning the parameters characterizing the distribution of nanopillar widths on the wavelength-dependent transmissivity of the random metasurface in the 400 – 800 nm wavelength range. This analysis helps us create a direct mapping between the parameters of the nanopillar width distribution and the spectral responses of the random metasurfaces. We exploit this mapping to design photonic devices encoding spectrally encrypted image data in the visible wavelength range. Our findings offer new insight into the optical properties of random media and provide avenues for developing such systems for a broad range of applications.