Skip to content

A reconfigurable multi-task diffractive optical processor using phase-multiplexed illumination

Oct 2026 · Emerging Topics in Artificial Intelligence (ETAI) 2026 · 0 citations

Abstract

We present a monochrome multi-task diffractive network framework that leverages illumination phase multiplexing to dynamically reconfigure its output transformations. Our design employs unique "phase keys"—distinct 2D illumination phase profiles—to encode specific transformations. The system activates a specific desired transformation simply by adjusting the input illumination phase, enabling programmable processing without physical hardware modifications. We demonstrate that a diffractive network composed of N=2TN_i N_o optimized diffractive features can realize T distinct complex-valued linear transformations – accurately executed for any complex field at the input aperture, where N_i and N_o refer to the input/output pixels, respectively. Proof-of-concept numerical analyses confirmed this phase-multiplexing capability, where T=512 distinct complex-valued transformations were implemented by the same monochrome diffractive network with negligible error. This framework offers a versatile, high-fidelity solution for programmable optical computing, with potential applications in adaptive machine vision and optical information processing.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.