Sandesh Kulkarni,
Yashraj Lade,
Sarvesh Mahimkar,
Anuj Mane,
Vaibhav Godase,
- Student, Department of Electronics and Telecommunication Engineering, SKN Sinhgad College of Engineering, Pandharpur, Maharashtra, India
- Student, Department of Electronics and Telecommunication Engineering, SKN Sinhgad College of Engineering, Pandharpur, Maharashtra, India
- Student, Department of Electronics and Telecommunication Engineering, SKN Sinhgad College of Engineering, Pandharpur, Maharashtra, India
- Student, Department of Electronics and Telecommunication Engineering, SKN Sinhgad College of Engineering, Pandharpur, Maharashtra, India
- Assistant Professor, Department of Electronics and Telecommunication Engineering, SKN Sinhgad College of Engineering, Pandharpur, Maharashtra, India
Abstract
Optical fibers provide high-bandwidth and low-loss transmission for modern communication networks; however, attenuation and chromatic dispersion remain two important physical limitations affecting transmission distance, pulse integrity, and system capacity. Conventional optical-fiber design generally relies on analytical calculations and repeated numerical simulations to determine suitable geometric and material parameters. Such approaches become computationally demanding when several fiber parameters are simultaneously optimized. This paper presents a machine-learning-based methodology for predicting and optimizing optical-fiber parameters to minimize attenuation and chromatic dispersion. A simulation-driven dataset containing 1200 optical-fiber configurations is considered by varying core radius, cladding radius, numerical aperture, refractive-index difference, operating wavelength, material attenuation coefficient, and fiber length. The corresponding attenuation coefficient and chromatic-dispersion coefficient are considered as output parameters. Random forest, support vector regression, artificial neural network, and extreme gradient boosting models are developed and compared using R^2, RMSE and MAE. An AI-assisted multi-objective optimization procedure is subsequently used to identify a fiber configuration with reduced attenuation and dispersion. The illustrative results indicate that XGBoost provides the highest prediction accuracy, with R^2values of 0.96 for attenuation and 0.95 for chromatic dispersion prediction. The optimized representative configuration provides an illustrative attenuation of 0.19 dB/km and chromatic dispersion of approximately 3.8 ps/(nm·km) at 1550 nm. The proposed methodology provides a computationally efficient framework for optical-fiber parameter optimization and can be extended to dispersion-shifted fibers, photonic crystal fibers, wavelength-division multiplexing systems, and next-generation high-capacity optical communication networks.
Keywords: Optical fiber, machine learning, attenuation, chromatic dispersion, fiber optimization, XGBoost, artificial neural network, optical communication, 1550 nm, WDM
[This article belongs to Trends in Opto-electro & Optical Communication ]
References
- Han J, Mesgarpour M, Asirvatham LG, Wongwises S, Ahn HS, Mahian O. A hyper-optimisation method based on a physics-informed machine learning and point clouds for a flat plate solar collector. J Therm Anal Calorim. 2023;148(13):6223-6242. doi:10.1007/s10973-023-12148-7.
- Madhankumar S, Ilangovan D, Viswanathan K. Integrating computational models and machine learning for corrugated absorber plate solar collector thermal predictions. Process Saf Environ Prot. 2024;188:336-349. doi:10.1016/j.psep.2024.05.069.
- Kumar R, Sharma N, Chahat GH, Ngo GH, Yadav AK, Lee D, et al. Prediction of jet impingement solar thermal air collector thermohydraulic performance using soft computing techniques. Case Stud Therm Eng. 2024;55:104144. doi:10.1016/j.csite.2024.104144.
- Ge B, Qi D, Zhou J, Qian N, Zhang L. Interpretable machine learning study of a collector based on combined twisted-tape and wavy-tape inserts. Case Stud Therm Eng. 2024;63:105236. doi:10.1016/j.csite.2024.105236.
- Gharaee H, Erfanimatin M, Bahman AM. Machine learning development to predict the electrical efficiency of photovoltaic-thermal (PVT) collector systems. Energy Convers Manag. 2024;315:118808. doi:10.1016/j.enconman.2024.118808.
- Sadeghzadeh M, Ahmadi MH, Kahani M, Sakhaeinia H, Chaji H, Chen L. Smart modeling by using artificial intelligent techniques on thermal performance of flat-plate solar collector using nanofluid. Energy Sci Eng. 2019;7(5):1649-1658. doi:10.1002/ese3.381.
- Said Z, Sharma P, Sundar LS, Li C, Tran DC, Khoa Pham ND, et al. Improving the thermal efficiency of a solar flat plate collector using MWCNT-Fe3O4/water hybrid nanofluids and ensemble machine learning. Case Stud Therm Eng. 2022;40:102448. doi:10.1016/j.csite.2022.102448.
- Jeon HS, Yeon SH, Park JK, Kim MH, Yoon Y, Kim CH, et al. ANN based solar thermal energy forecasting model and its heating energy saving effect through thermal storage. Appl Therm Eng. 2025;267:125740. doi:10.1016/j.applthermaleng.2025.125740.
- Du B. Improving the accuracy of machine learning models in predicting evacuated tube solar collector through Random Forest replacement. Therm Sci Eng Prog. 2026;69:104480. doi:10.1016/j.tsep.2026.104480.
- Hosouli S, Jackson-French L, Pierucci G, De Lucia M. Predicting outdoor performance of solar parabolic trough collector (PTC) from indoor test data using machine learning. Appl Therm Eng. 2026;290:130020. doi:10.1016/j.applthermaleng.2026.130020.
- Assareh E, Riaz A, Ahmadinejad M, Hoseinzadeh S, Abdehvand MZ, Keykhah S, et al. Enhancing solar thermal collector systems for hot water production through machine learning-driven multi-objective optimization with phase change material (PCM). J Energy Storage. 2023;73:108990. doi:10.1016/j.est.2023.108990.
- Godase V. Cross-domain comparative analysis of microwave imaging systems for medical diagnostics and industrial testing. J Microw Eng Technol. 2025;12(2):39-48.
- Jamadade VK, Ghodke MG, Katakdhond SS, Godase V. A review on real-time substation feeder power line monitoring and auditing systems. Int J Emerg IoT Technol Smart Electron Commun. 2025;1(2):1-6.
- Godase VV. VLSI-integrated energy harvesting architectures for battery-free IoT edge systems. J Electron Des Technol. 2025;2:1-2.
- Godase V, Pawar P, Nagane S, Kumbhar S. Automatic railway horn system using NodeMCU. J Control Instrum. 2024;15(1):11-19.
Trends in Opto-electro & Optical Communication
| Volume | 16 | |
| Issue | 02 | |
| Received | 21/09/2026 | |
| Accepted | 22/09/2026 | |
| Published | 23/09/2026 | |
| Publication Time | 2 Days |