Silicon nitride nanophotonics for low-redundancy 3D convolution
Abstract
Three-dimensional (3D) convolution extracts correlations in high-dimensional data, but overlapping receptive fields introduce substantial redundant data movement. Here, we demonstrate a silicon nitride 3D optical convolution accelerator (3D-OCA) that reconstructs receptive fields through coordinate-aware wavelength-to-group-delay mapping. A deterministically serialized input tensor is broadcast onto multiple wavelength channels, and a chirped waveguide Bragg grating (CWBG) compensates the temporal offsets associated with 3D kernel coordinates. This arrangement continuously forms neighboring receptive fields without repeatedly rearranging and loading their shared input samples. The integrated CWBG provides a differential group delay of 2691 ps and a dispersion of 124.8 ps/nm. At 20 Gbaud, spatial-spectral processing of Indian Pines data yields convolution agreement with a coefficient of determination up to 0.997 and 97.9% classification accuracy, compared with 98.9% digitally. At 10 Gbaud, the optical convolution layer preserves spatiotemporal features and achieves 92.5% accuracy on a four-class KTH video-recognition task. These results establish low-redundancy streaming 3D convolution across spectral and temporal data dimensions using the same optical delay architecture.
Related
- Single-step initialization of a 16-channel silicon nitride DWDM demultiplexer using a physics-informed digital twin
- Compact thermo-optic switch based on graphene-silicon nitride nanobeam cavity
- Broadband Visible-Spectrum Single-Mode Integrated Photonics on a Silicon Nitride Platform
- Thermo-optically tunable Bragg soliton-effect compression in on-chip silicon nitride gratings
- An Ultracompact RGB (de)multiplexer based on a Silicon Nitride dual-segmented MMI
BibTeX
@misc{hu2026silicon,
title = {{Silicon nitride nanophotonics for low-redundancy 3D convolution}},
author = {Shuai Hu and Jingcheng Li and Qichao Ding and Hongli Wang and Hailong Zhou and Chi Zhang and Jianji Dong and Xinliang Zhang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.37246},
eprint = {2609.37246},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.37246},
}
RIS
TY - UNPB
TI - Silicon nitride nanophotonics for low-redundancy 3D convolution
AU - Shuai Hu
AU - Jingcheng Li
AU - Qichao Ding
AU - Hongli Wang
AU - Hailong Zhou
AU - Chi Zhang
AU - Jianji Dong
AU - Xinliang Zhang
PY - 2026
DA - 2026/09/29
PB - arXiv
DO - 10.48550/arxiv.2609.37246
UR - https://arxiv.org/abs/2609.37246
AB - Three-dimensional (3D) convolution extracts correlations in high-dimensional data, but overlapping receptive fields introduce substantial redundant data movement. Here, we demonstrate a silicon nitride 3D optical convolution accelerator (3D-OCA) that reconstructs receptive fields through coordinate-aware wavelength-to-group-delay mapping. A deterministically serialized input tensor is broadcast onto multiple wavelength channels, and a chirped waveguide Bragg grating (CWBG) compensates the temporal offsets associated with 3D kernel coordinates. This arrangement continuously forms neighboring receptive fields without repeatedly rearranging and loading their shared input samples. The integrated CWBG provides a differential group delay of 2691 ps and a dispersion of 124.8 ps/nm. At 20 Gbaud, spatial-spectral processing of Indian Pines data yields convolution agreement with a coefficient of determination up to 0.997 and 97.9% classification accuracy, compared with 98.9% digitally. At 10 Gbaud, the optical convolution layer preserves spatiotemporal features and achieves 92.5% accuracy on a four-class KTH video-recognition task. These results establish low-redundancy streaming 3D convolution across spectral and temporal data dimensions using the same optical delay architecture.
ER -