Reconfigurable beam shaping with rotating phase masks using diffractive neural networks

Computational Optics, pp. 12

Abstract

Diffractive neural networks (DNNs) as physical representations of artificial neural networks have demonstrated remarkably flexible potential as optical systems for laser beam shaping but dynamic reconfiguration typically relies on active spatial light modulators (SLMs). We present a method to train DNNs so that rotating one or more passive phase masks switches the device among multiple, completely distinct beam-shaping functionalities, enabling dynamic beam control without SLMs. The method can also be realized experimentally with reflective DOEs produced with a unique direct laser writing approach.

Authors 7

1. Paul Buske Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen Univ. (Germany)

2. Louis Michels Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen Univ. (Germany)

3. Dominik Onyszkiewicz Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen Univ. (Germany)

4. Christian Wahl
   Affiliation as printed

   Midel Photonics GmbH (Germany)

5. Christopher Grossert
   Affiliation as printed

   Midel Photonics GmbH (Germany)

6. Annika Bonhoff Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen Univ. (Germany)

7. Carlo Holly Aachen Fraunhofer-Institut für Lasertechnik ILT

   Fraunhofer Institute for Laser Technology · RWTH Aachen University

   Affiliation as printed

   Fraunhofer-Institut für Lasertechnik ILT (Germany)

   RWTH Aachen Univ. (Germany)