Design of ultracompact broadband focusing spectrometers based on deep diffractive neural networks

TL;DR

This paper introduces a novel inverse design method for ultracompact broadband focusing spectrometers using deep diffractive neural networks, enabling efficient spectral analysis in a compact form.

Contribution

The work presents the first systematic design of two-layer diffractive devices with engineered angular dispersion for broadband focusing and spectral reconstruction using deep neural networks.

Findings

Achieved 5 nm spectral resolution across the visible spectrum.

Demonstrated accurate spectrum reconstruction from a superluminescent diode.

Designed devices with 100 μm side length and 300 μm focal length.

Abstract

We propose the inverse design of ultracompact, broadband focusing spectrometers based on adaptive deep diffractive neural networks (a-D$^2$NNs). Specifically, we introduce and characterize two-layer diffractive devices with engineered angular dispersion that focus and steer broadband incident radiation along predefined focal trajectories with desired bandwidth and $5$ nm spectral resolution. Moreover, we systematically study the focusing efficiency of two-layer devices with side length $L=100~\mu\mathrm{m}$ and focal length $f=300~\,\mu\mathrm{m}$ across the visible spectrum and we demonstrate accurate reconstruction of the emission spectrum from a commercial superluminescent diode. The proposed a-D$^2$NNs design method extends the capabilities of efficient multi-focal diffractive optical devices to include single-shot focusing spectrometers with customized focal trajectories for applications to ultracompact multispectral imaging and lensless microscopy.