Miniaturized Spectrometer Enabled by End-to-End Deep Learning on Large-Scale Radiative Cavity Array

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, a mini-spectrometer that combines a large-scale radiative cavity array with end-to-end deep learning networks is proposed and demonstrated. Specifically, high-Q bound states are utilized in continuum cavities with distinct radiation characteristics as the fundamental units to achieve parallel spectral detection. An array of resonators with quality factors above 10⁴ over a wide spectral range, from 1525 to 1605 nm, is realized. A deep network with 8000 outputs is further trained 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.