Towards integrated mode-division demultiplexing spectrometer by deep learning

Abstract

Miniaturized spectrometers have been widely researched in recent years, but few studies are conducted with on-chip multimode schemes for mode-division multiplexing (MDM) systems. Here we propose an ultracompact mode-division demultiplexing spectrometer that includes branched waveguide structures and graphene-based photodetectors, which realizes simultaneously spectral dispersing and light fields detecting. In the bandwidth of 1500-1600 nm, the designed spectrometer achieves the single-mode spectral resolution of 7 nm for each mode of TE 1 -TE 4 by Tikhonov regularization optimization. Empowered by deep learning algorithms, the 15-nm resolution of parallel reconstruction for TE 1 -TE 4 is achieved by a single-shot measurement. Moreover, by stacking the multimode response in TE 1 -TE 4 to the single spectra, the 3-nm spectral resolution is realized. This design reveals an effective solution for on-chip MDM spectroscopy, and may find applications in multimode sensing, interconnecting and processing.

Optimized design for absorption metasurface based on autoencoder (AE) and BiLSTM-Attention-FCN-Net

“…Deep learning uses neural networks to learn patterns in data, and after training and optimizing on a dataset of metasurfaces, neural networks can effectively predict the best metasurface design. The design method of using deep learning to optimize the metasurface reduces the design cycle time and cost, and improves the design accuracy and efficiency for the metasurface [40][41][42][43]. However, the existing methods mainly use the full connection and the convolutional neural network.…”

Enhancing detectivity in mid-infrared photodetectors through structural parameter engineering in HgSe-HgTe colloidal quantum dots

“…1 The spectrophotometers typically consist of two main components: wavelength-selective elements and photodetection elements. 2,3 Photodetection technologies, which convert photons into electrical signals, have been extensively investigated. 4 These photodetectors are categorized based on the range of electromagnetic radiation absorbed by the active layer, including X-ray, visible, and infrared (IR) photodetectors.…”

“…The interaction of light and matter has served as the foundation for optoelectronic devices and remains a focal point in quantum science 1 . The spectrophotometers typically consist of two main components: wavelength-selective elements and photodetection elements 2 , 3 . Photodetection technologies, which convert photons into electrical signals, have been extensively investigated 4 .…”

Planar peristrophic multiplexing metasurfaces

“…Note that the degree of freedom (DoF) is the cornerstone of multiplexing methods. Nowadays, a great number of impressive efforts have been devoted to dividing various physical properties as the discriminating DoFs, including wavelength 17 , polarization 18 , and spatial modes 19 . Moreover, a nontrivial phase mode, named orbital angular momentum (OAM), has been recently exploited for planar multiplexing 20−23 .…”