Miniaturized spectrometer enabled by end-to-end deep learning on large-scale radiative cavity array

TL;DR

This paper introduces a miniaturized spectrometer that uses a large-scale radiative cavity array combined with deep learning to achieve high-resolution, broadband spectral detection in a compact device.

Contribution

The work presents a novel integration of a large-scale cavity array with end-to-end deep learning for mini-spectrometry, enabling high resolution and broad spectral range in a small form factor.

Findings

Achieved 0.048 nm spectral resolution within 80 nm bandwidth.

Demonstrated over 95% spectral fidelity in experimental tests.

Implemented a 36x30 cavity array spanning 1525-1605 nm spectral range.

Abstract

Miniaturized (mini-) spectrometers are highly desirable tools for chemical, biological, and medical diagnostics because of their potential for portable and in situ spectral detection. In this work, we propose and demonstrate a mini-spectrometer that combines a large-scale radiative cavity array with end-to-end deep learning networks. Specifically, we utilize high-Q bound states in continuum cavities with distinct radiation characteristics as the fundamental units to achieve parallel spectral detection. We realize a 36 $\times$ 30 cavity array that spans a wide spectral range from 1525 to 1605 nm with quality factors above 10^4. We further train a deep network with 8000 outputs to directly map arbitrary spectra to array responses excited by the out-of-plane incident. Experimental results demonstrate that the proposed mini-spectrometer can resolve unknown spectra with a resolution of 0.048 nm in a bandwidth of 80 nm and fidelity exceeding 95%, thus offering a promising method for compact, high resolution, and broadband spectroscopy.