Software-Defined Reconfigurable Underwater Wireless Optical Communication System With Laser Array Modulation and Deep Learning Demodulation
Journal of Lightwave Technology · 15.06.2026 · 44 (12) · 4937-4945
DOI: 10.1109/jlt.2026.3685456 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Lightwave Technology, erschienen am 15.06.2026. Verfasst von Tianchen Tong und 5 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Beijing Municipality, China Postdoctoral Science Foundation.
Abstract (Originalsprache)
Underwater wireless optical communication (UWOC) systems face an important trade-off between transmission rate and link robustness, which is further exacerbated by channel attenuation and the nonlinearity of opto-electronic conversion. To address these challenges, we propose a software-defined reconfigurable UWOC (SDR-UWOC) system based on laser diode (LD) arrays for optical superimposition without complex electrical digital-to-analog converters (DACs). The system employs a unified hardware platform that supports dynamic modulation and data rate switching between power-superimposed on-off keying (OOK) for robust communication and signal-superimposed pulse amplitude modulation with four levels (PAM4) for high-speed transmission. To mitigate optical device nonlinearities and channel impairments, we develop a hybrid convolutional neural network and long short-term memory (CNN-LSTM) demodulation scheme that combines spatial feature extraction with temporal sequence modeling. Experimental validation over a 1-meter water channel with controlled attenuation demonstrates that OOK superimposition improves attenuation tolerance by 3.1 dB compared to single-LD operation. In PAM4 mode, the optical combination of three LDs extends the transmitter's linear operating region by approximately 2.4-fold and the CNN-LSTM demodulation algorithm achieves a 0.8 dB improvement in attenuation tolerance compared to fixed-threshold decisions. Through software-defined reconfiguration between each modulation mode, the system achieves a 6.9 dB improvement in attenuation tolerance and extends the usable transmission rate range seven-fold (100-700 Mbps) compared to fixed-modulation baselines. This work provides a feasible solution for dynamically balancing data rate and link robustness in complex underwater optical communication networks, showing promising improvements in both link performance and transmission rate flexibility.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Beijing Municipality
- China Postdoctoral Science Foundation
Zitieren
APA
Tong, T., Gao, G., Zhang, J., Ma, Z., Guo, Y., & Huang, S. (2026). Software-Defined Reconfigurable Underwater Wireless Optical Communication System With Laser Array Modulation and Deep Learning Demodulation. Journal of Lightwave Technology, 44(12), 4937-4945. https://doi.org/10.1109/jlt.2026.3685456
Vancouver
Tong T, Gao G, Zhang J, Ma Z, Guo Y, Huang S. Software-Defined Reconfigurable Underwater Wireless Optical Communication System With Laser Array Modulation and Deep Learning Demodulation. Journal of Lightwave Technology. 2026;44(12):4937-4945. doi:10.1109/jlt.2026.3685456
BibTeX
@article{Tong2026,
author = {Tong, Tianchen and Gao, Guanjun and Zhang, Jialiang and Ma, Ziqi and Guo, Yonggang and Huang, Shanguo},
title = {Software-Defined Reconfigurable Underwater Wireless Optical Communication System With Laser Array Modulation and Deep Learning Demodulation},
journal = {Journal of Lightwave Technology},
year = {2026},
volume = {44},
number = {12},
pages = {4937--4945},
doi = {10.1109/jlt.2026.3685456}
}
EUVIMED