Light features three tunable variables: amplitude, phase and polarization. Harnessing all three simultaneously for optical modulation and demodulation has been a long-standing cross-century challenge in the information era. Recently, Professor Tan Xiaodi’s team from the College of Photonic and Electronic Engineering, Fujian Normal University, has pioneered an encoding-decoding method for concurrent modulation of the three optical variables. The work breaks the conventional limit of modulating at most two variables and greatly boosts holographic data storage capacity. The paper titled *Encoding and decoding of multidimensional optical field modulation in holographic data storage* was published in *Optica*.

This technical breakthrough builds on the team’s long-term research into tensor polarization holography and flexible deployment of deep learning. The theory enables faithful recording and error-free reconstruction of all three optical variables — a prerequisite to avoid bit errors during simultaneous modulation and demodulation. With this foundation, polarization can be incorporated as an independent variable into the coding framework to construct multidimensional information encoding schemes.

For encoding, the team leverages dual-phase holography and a single phase-only spatial light modulator to realize coordinated modulation of optical amplitude, phase and polarization. For decoding, a dual-input, triple-output deep neural network named TriDecode-Net is designed. It reconstructs amplitude, phase and polarization from two complementary diffracted intensity images to achieve synchronous multidimensional decoding. When applied to holographic storage, the method substantially raises storage density and data throughput. The technology can also be extended to other light-based information processing applications.

Schematic: Holographic data storage strategy with multidimensional information modulation

Fujian Normal University serves as the primary affiliation. PhD candidates Chen Ruixian and Wang Jinyu are co-first authors. Professor Tan Xiaodi and Associate Professor Lin Dakui are co-corresponding authors. The research was supported by the Joint Fund of the National Natural Science Foundation of China and other grants.