Comparative Study of ML and DL in Optical Transceiver Failure Diagnosis
Article Sidebar
Main Article Content
Abstract
High-speed optical transceivers require robust failure analysis methods to ensure production reliability in modern communication systems. This study systematically evaluates machine learning algorithms (Random Forest, XGBoost) and deep learning approaches (Fully Connected Neural Networks) for optical transceiver failure analysis across two operational scenarios using real manufacturing data from 6,446 units. In a comprehensive data analysis (Scenario #1), both Random Forest and XGBoost achieved exceptional performance (MSE: 0.0000, MAE: 0.0001), while FCNN demonstrated comparable results (Loss: 0.0002, MAE: 0.0002). In a focused analysis of failed units (Scenario #2), XGBoost outperformed other models with the lowest error metrics (MSE: 0.0091, MAE: 0.0165) compared to Random Forest (MSE: 0.0125, MAE: 0.0399) and FCNN (Loss: 0.1571, MAE: 0.2987). SHAP analysis consistently identifies influential features across both scenarios, providing actionable insights for quality control optimization. These findings establish a quantitative framework for selecting optimal AI approaches for optical transceiver failure diagnostics, with machine learning models recommended for datasets under 10,000 samples and deep learning for larger datasets. The proposed methodology advances AI-driven failure diagnostics in optical transceiver manufacturing.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
I/we certify that I/we have participated sufficiently in the intellectual content, conception and design of this work or the analysis and interpretation of the data (when applicable), as well as the writing of the manuscript, to take public responsibility for it and have agreed to have my/our name listed as a contributor. I/we believe the manuscript represents valid work. Neither this manuscript nor one with substantially similar content under my/our authorship has been published or is being considered for publication elsewhere, except as described in the covering letter. I/we certify that all the data collected during the study is presented in this manuscript and no data from the study has been or will be published separately. I/we attest that, if requested by the editors, I/we will provide the data/information or will cooperate fully in obtaining and providing the data/information on which the manuscript is based, for examination by the editors or their assignees. Financial interests, direct or indirect, that exist or may be perceived to exist for individual contributors in connection with the content of this paper have been disclosed in the cover letter. Sources of outside support of the project are named in the cover letter.
I/We hereby transfer(s), assign(s), or otherwise convey(s) all copyright ownership, including any and all rights incidental thereto, exclusively to the Journal, in the event that such work is published by the Journal. The Journal shall own the work, including 1) copyright; 2) the right to grant permission to republish the article in whole or in part, with or without fee; 3) the right to produce preprints or reprints and translate into languages other than English for sale or free distribution; and 4) the right to republish the work in a collection of articles in any other mechanical or electronic format.
We give the rights to the corresponding author to make necessary changes as per the request of the journal, do the rest of the correspondence on our behalf and he/she will act as the guarantor for the manuscript on our behalf.
All persons who have made substantial contributions to the work reported in the manuscript, but who are not contributors, are named in the Acknowledgment and have given me/us their written permission to be named. If I/we do not include an Acknowledgment that means I/we have not received substantial contributions from non-contributors and no contributor has been omitted.
References
L. M. Choong and W. Kuang, "Advanced Data Analytics and Supervised Machine Learning in Optics Engineering Analysis," in Proc. 2021 Int. Conf. Computer & Information Sciences (ICCOINS), 2021.
F. Boeuf, A. Fincato, and L. Maggi, "A Silicon Photonics Technology for 400 Gbit/s Applications," in Proc. 2022 Int. Symp. VLSI Design, Automation and Test (VLSI-DAT), 2022.
P. Dong, J. Chen, and A. Melikyan, "Silicon Photonics for 800G and Beyond [invited]," in Proc. 2022 Optical Fiber Communications Conf. and Exhibition (OFC), 2022.
B. Chen, Y. Zhang, and Y. Zhu, "Trends in Optoelectronic IC for Recent Optical Module and Photonics Integration," in Proc. 2020 IEEE Int. Conf. Integrated Circuits, Technologies and Applications (ICTA), 2020.
H. Li, L. Luo, and Q. Chen, "800G DR8 Transceiver Based on Thin-film Lithium Niobate Photonic Integrated Circuits," in Proc. 2022 European Conf. Optical Communication (ECOC), 2022.
P. Xiang and X. Song, "Research and design of 800Gbit/s PAM4 LR8 10km optical module," in Proc. 2022 7th Int. Conf. Communication, Image and Signal Processing (CCISP), 2022.
D. Coenen, H. Oprins, and I. De Wolf, "Circuit-level Thermal Modelling of Silicon Photonic Transceiver Array using Machine Learning," in Proc. 2022 21st IEEE Intersociety Conf. Thermal and Thermomechanical Phenomena in Electronic Systems (iTherm), 2022.
L. M. Choong and W. K. Cheng, "Machine Learning in Failure Analysis of Optical Transceiver Manufacturing Process," in Proc. 2021 Int. Conf. Computer & Information Sciences (ICCOINS), 2021.
M. Valvo, E. Andrianopoulos, and P. Groumas, "Optical bench for multi-lane high-rate transceiver testing," in Proc. 2022 Italian Conf. Optics and Photonics (ICOP), 2022.
D. Lippiatt, A. Melgar, and V. A. Thomas, "Impairment Identification for PAM-4 Transceivers and Links Using Machine Learning," in Proc. 2021 Optical Fiber Communications Conf. (OFC), 2021.
C. Wang, L. Wang, and Z. Wang, "Data, Analytics and Unsupervised Learning Enabled Proactive Maintenance for Optical Transceivers in Hyperscale Data Centers," in Proc. 2021 Opto-Electronics and Communications Conf. (OECC), 2021.
K. Abdelli, C. Tropschug, H. Grießer, and S. Pachnicke, "Machine-learning-based anomaly detection in optical fiber monitoring," J. Opt. Commun. Netw., vol. 14, no. 5, pp. 365-375, 2022.
O. Nyarko-Boateng, A. F. Adekoya, and B. A. Weyori, "Using machine learning techniques to predict the cost of repairing hard failures in underground fiber optics networks," J. Big Data, vol. 7, Art. 64, 2020.
D. Wang, C. Zhang, W. Chen, H. Yang, M. Zhang, and A. P. T. Lau, "A review of machine learning-based failure management in optical networks," Sci. China Inf. Sci., vol. 65, no. 11, Art. 211302, 2022.
M. Li, D. Wang, and Q. Cui, "End-to-End Deep Learning of Optical Fiber," in Proc. 2020 Opto-Electronics and Communications Conf. (OECC), 2020.
F. Musumeci, "Machine Learning for Failure Management in Optical Networks," in Proc. 2021 Optical Fiber Communication Conf. (OFC), paper Th4J.1, 2021.
L. E. Kruse, S. Kühl, A. Dochhan, and S. Pachnicke, "Experimental investigation of machine-learning-based soft-failure management using the optical spectrum," J. Opt. Commun. Netw., vol. 16, no. 2, pp. 94-103, 2024.
S. Behera, T. Panayiotou, and G. Ellinas, "Machine learning framework for timely soft-failure detection and localization in elastic optical networks," J. Opt. Commun. Netw., vol. 15, no. 10, pp. E74-E85, 2023.
J. Krzyston, R. Bhattacharjea, and A. Stark, "Modulation Pattern Detection Using Complex Convolutions in Deep Learning Communications," in Proc. 2020 25th Int. Conf. Pattern Recognition (ICPR), 2020.
H. Lun et al., "A GAN based soft failure detection and identification framework for long-haul coherent optical communication systems," J. Lightw. Technol., vol. 41, pp. 2312-2322, 2023.
J. Xiang et al., "Low-complexity conditional generative adversarial network (c-GAN) based nonlinear equalizer for coherent data-center interconnections," J. Lightw. Technol., vol. 41, pp. 5966-5972, 2023.
J. Li, D. Wang, M. Zhang, and S. Cui, "Digital twin-enabled self-evolved optical transceiver using deep reinforcement learning," Opt. Lett., vol. 45, pp. 4654-4657, 2020.
Z. Sun, C. Zhang, M. Zhang, B. Ye, and D. Wang, "Expertise-Embedded Machine Learning for Enhanced Failure Management of Optical Modules in OTN," in Proc. 2024 Optical Fiber Communication Conf. (OFC), paper Th3I.5, 2024.
J. Babbar, A. Triki, R. Ayassi, and M. Laye, "Machine learning models for alarm classification and failure localization in optical transport networks," J. Opt. Commun. Netw., vol. 14, no. 8, pp. 621-628, 2022.
K. K. Soothar, Y. Chen, A. H. Magsi, C. Hu, and H. Shah, "Optimizing Optical Fiber Faults Detection: A Comparative Analysis of Advanced Machine Learning Approaches," Comput., Mater. Continua, vol. 79, no. 2, pp. 2697-2721, 2024.
L. Zhang, L. Yan, W. Shen, F. Li, J. Wu, and W. Liang, "Neural network-based fiber optic cable fault prediction study for power distribution communication network," Appl. Math. Nonlinear Sci., vol. 9, no. 1, 2023.
L. Z. Khan, J. Pedro, N. Costa, A. Napoli, and N. Sambo, "Data Augmentation to Improve Machine Learning for Optical Network Failure Management," in Proc. 2022 European Conf. and Exhibition on Optical Communication (ECOC), paper We4B.3, 2022.
H. Song, R. Lin, L. Wosinska, P. Monti, M. Zhang, Y. Liang, Y. Li, and J. Zhang, "Cluster-based unsupervised method for eavesdropping detection and localization in WDM systems," J. Opt. Commun. Netw., vol. 16, no. 10, pp. F52-F61, 2024.
A. A. Kashtanov, M. E. Pazhetnov, and E. V. Kashtanova, "Development of a Machine Learning Based Algorithm to Identify the Main Types of Optical Signal Transmission Faults Between a Switch and an Optical Transceiver," in Proc. 2021 XV Int. Scientific-Technical Conf. Actual Problems of Electronic Instrument Engineering (APEIE), 2021.
M. T. Ribeiro, S. Singh, and C. Guestrin, "Why Should I Trust You?: Explaining the Predictions of Any Classifier," in Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., 2016, pp. 1135–1144.
