Leveraging statistical-spectral correlations of random metasurfaces for steganography and multi-wavelength cryptography

Abstract

The ability to tailor the spectral response of photonic devices is paramount to the advancement of a broad range of applications. The vast design space offered by disordered optical media provides enhanced functionality for spectral tailoring while also making it challenging to map the spectral properties of such complex systems to their structural attributes. In this work, we investigate correlations between the configuration statistics of random metasurfaces and their spectral transmissivity in the visible, and leverage those to develop a reduced phase space. In the latter part of the manuscript, we use this reduced phase space to design a pixelated color filter that hides visual data within a preselected cover image for steganography. Furthermore, we design a pair of color filters that can collectively encrypt a given grayscale image in their spectral transmissivities. We envision such devices to create opportunities for the development of compact, next-generation cryptographic systems. More broadly, the results presented in this manuscript provide new, to the best of our knowledge, avenues for optimizing large-scale random metasurfaces to achieve enhanced optical functionalities for a wide variety of applications.