https://doi.org/10.1140/epjd/s10053-026-01205-z
Research - Photon
Defect-immune topological transmission in C4-symmetric photonic crystals via neural network
School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, 516 Jungong Road, Shanghai, China
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
23
November
2025
Accepted:
8
June
2026
Published online:
1
July
2026
Abstract
Defect-immune transmission in topological photonic crystals is essential for developing highly stable optical information systems. In this study, we present a C₄-symmetric topological photonic crystal integrated with a fully connected neural network to achieve robust transmission. By systematically tuning the spatial configuration of cylindrical elements within the unit cell, we facilitate controllable topological phase transitions and observe topologically protected edge and corner states at interfaces between distinct topological phases. Furthermore, a deep learning model is established to predict the influence of structural deformation defects—specifically, the transformation from cylindrical pillars to elliptical ones—on transmission properties, achieving a prediction error on the order of 10⁻3. Based on this model, we design and demonstrate a three-channel wavelength-division multiplexer capable of selectively transmitting multiple optical signals while preserving robustness under complex defect perturbations. This work demonstrates a comprehensive data-driven framework for constructing highly stable and defect-tolerant optical transmission systems, which is potential of combining topological photonics with machine learning to advance optical communications.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, SIF and Springer-Verlag GmbH Germany, part of Springer Nature 2026
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
