Integrated Polarization, Distance, and Rotation for Multi-DoF Diffractive Processor and Information Encryption

All-optical diffractive deep neural networks (D²NNs) offer significant advantages in processing speed and power consumption, thereby accelerating the development of optical computing and artificial intelligence (AI). Integrating multiple degrees of freedom (multi-DoF) into D²NNs is a pivotal role in improving information processing and task-loading capacity, an enormous challenge in current all-optical diffractive computing/processors. Here, a multi-DoF diffractive processor is proposed and experimentally demonstrated that leverages a metasurfaces-based approach to integrate polarization, distance, and rotation channels for versatile inference tasks and information encryption. The approach is validated using three-layer metasurfaces that enable high task-capacity tasks, including single-/dual-digit and single-/dual-fashion-product classification, logic operators, and image transformation. Moreover, by mapping large volumes of input data into multi-DoF channels and encoding the information in Morse code with our D²NNs framework, a high-security information transmission system is experimentally implemented. The integration of polarization, distance, and rotation channels into an all-optical diffractive processor with multifunctional capabilities paves the way for multifunctional integrated devices and communication.