Comparative Analysis of Random Forest and Long Short-Term Memory for Predicting Optical Power Degradation in FTTH Networks
DOI:
https://doi.org/10.30871/jaic.v10i4.13717Keywords:
FTTH, Long Short-Term Memory, optical power degradation, predictive maintenance, Random ForestAbstract
Fiber-to-the-Home (FTTH) networks are widely used to provide high-speed broadband services, but optical power degradation can reduce network performance and service quality. This study compares Random Forest (RF) and Long Short-Term Memory (LSTM) for predicting FTTH network conditions classified as Normal, Warning, and Critical. The study used 63,145 historical records collected from 58 Optical Network Terminals (ONTs) between March and May 2026. To provide a fair comparison, RF was trained using engineered tabular features, including lag and rolling-window statistics, while LSTM used six-step sequential data representing approximately the previous six hours. Model performance was evaluated using accuracy, precision, recall, F1-score, Mean Absolute Error (MAE), Root Mean Square Error (RMSE), training time, and inference time. The results show that RF substantially outperformed LSTM, achieving 98.41% accuracy, precision, recall, and F1-score, with an MAE of 0.0173 and RMSE of 0.1420. RF also required only 2.667 seconds for training and 0.0102 ms for inference, compared with 189.98 seconds and 0.1856 ms for LSTM. Per-class evaluation confirmed that RF performed well across all network conditions, with precision and recall above 99% for Normal, above 94% for Warning, and above 90% for Critical. A strict chronological train-test split further confirmed the robustness of RF, which achieved 98.59% accuracy. Feature importance analysis showed that historical optical power, particularly lag-based features, was the most influential predictor of network degradation. These findings indicate that FTTH optical power degradation can be effectively modeled using engineered tabular features rather than a purely sequential approach. Finally, the RF model was integrated into a web-based monitoring dashboard with WhatsApp-based early warnings to support proactive FTTH network maintenance.
Downloads
References
[1] H. Hermansyah, A. Khaidar, N. Nurdin, and S. Kurnia, “Implementation of Static Routing and Quality of Service for Optimization of Network Traffic Management on Cisco Routers,” J. Artif. Intell. Softw. Eng., vol. 5, no. 3, pp. 1222–1230, 2025.
[2] Y. Chen and M. Xu, “Predictive analytics for optical network degradation using ensemble learning,” IEEE Trans. Netw. Manag., vol. 33, no. 1, pp. 1–10, 2025.
[3] D. Prasetyo and A. Wibowo, “Prediksi kualitas sinyal optik menggunakan LSTM pada jaringan FTTH,” J. Telekomun. dan Komput., vol. 14, no. 2, pp. 101–109, 2024.
[4] N. Ahmad and A. Rahman, “Hybrid Random Forest and LSTM model for network performance prediction,” Int. J. Intell. Eng. Syst., vol. 17, no. 1, pp. 210–220, 2024.
[5] S. Liu and K. Zhang, “FTTH network challenges in supporting IoT and cloud services,” Comput. Networks, vol. 243, pp. 110–118, 2024.
[6] R. Singh and P. Kumar, “Deep learning approaches for predictive maintenance in optical networks,” IEEE Access, vol. 11, pp. 78900–78912, 2023.
[7] A. Putra and D. Nugroho, “Penerapan machine learning untuk prediksi gangguan jaringan optik,” J. RESTI, vol. 6, no. 3, pp. 450–458, 2022.
[8] L. Zhou and H. Chen, “Optical power prediction in FTTH networks using LSTM,” Opt. Switch. Netw., vol. 48, pp. 100–108, 2023.
[9] R. Hakim and F. Siregar, “Pendekatan pemeliharaan preventif jaringan FTTH berbasis data historis,” J. Inform., vol. 17, no. 1, pp. 55–63, 2023.
[10] X. Li and Y. Zhao, “Random Forest-based classification for network performance degradation,” Appl. Soft Comput., vol. 120, pp. 108–115, 2022.
[11] P. Almeida and L. Costa, “FTTH network monitoring and fault detection using machine learning,” IEEE Commun. Mag., vol. 59, no. 2, pp. 40–46, 2021.
[12] H. Wang and Q. Liu, “Analysis of optical signal attenuation in FTTH networks,” Optik (Stuttg)., vol. 242, pp. 167–174, 2021.
[13] B. Santoso and R. Pratama, “Analisis redaman jaringan FTTH berbasis serat optik,” J. Teknol. Inf. dan Komun., vol. 11, no. 2, pp. 85–92, 2022.
[14] M. Rashed, S. Ahmed, and M. Hossain, “Optical network performance analysis using Random Forest algorithm,” Opt. Fiber Technol., vol. 56, pp. 102–110, 2020.
[15] Y. Zhang and X. Wang, “LSTM-based network traffic prediction for optical networks,” IEEE Access, vol. 6, pp. 35612–35621, 2018.
[16] J. Kim and S. Cho, “Time-series performance prediction in communication networks using LSTM,” J. Netw. Comput. Appl., vol. 129, pp. 12–21, 2019.
[17] D. Prasetyo and E. Santoso, “Implementasi LSTM untuk prediksi data deret waktu,” J. Ilmu Komput., vol. 19, no. 1, pp. 44–52, 2024.
[18] A. Rahman and M. Hossain, “Backpropagation optimization techniques in deep learning,” Appl. Soft Comput., vol. 118, pp. 108–116, 2022.
[19] P. Singh and R. Kumar, “Loss functions in deep learning for regression problems,” Expert Syst. Appl., vol. 210, pp. 118–125, 2023.
[20] R. S. Sutton and A. G. Barto, “Reinforcement learning: An introduction and recent developments,” IEEE Trans. Neural Networks, vol. 31, no. 11, pp. 4485–4496, 2020.
[21] Y. Zhou and K. Chen, “Activation functions in deep neural networks: A review,” Neural Networks, vol. 158, pp. 95–110, 2023.
[22] A. Pratama and Y. Nugroho, “Analisis kualitas layanan jaringan FTTH berbasis Passive Optical Network,” J. Teknol. Telekomun., vol. 9, no. 1, pp. 15–24, 2024.
[23] M. Rahman and S. Hossain, “Performance evaluation of FTTH networks using GPON architecture,” Int. J. Electr. Comput. Eng., vol. 13, no. 4, pp. 3890–3898, 2023.
[24] D. Tsonev and H. Haas, “Fiber-to-the-Home networks for high-speed broadband access,” IEEE Commun. Mag., vol. 59, no. 3, pp. 62–68, 2021.
[25] J. Gao and Y. Liu, “Prediction and the influencing factor study of colorectal cancer hospitalization costs in China based on machine learning-random forest and support vector regression: a retrospective study,” Front Public Heal., vol. 12, no. 1211220, 2024, doi: 10.3389/fpubh.2024.1211220.
[26] Y. Zhou, H. Shen, and M. Zhang, “A Distributed and Privacy-Preserving Random Forest Evaluation Scheme with Fine Grained Access Control,” Symmetry (Basel)., vol. 14, no. 2, p. 415, 2022, doi: 10.3390/sym14020415.
[27] E. R. B. Sebayang, Y. H. Chrisnanto, and M. Melina, “Klasifikasi Data Kesehatan Mental di Industri Teknologi Menggunakan Algoritma Random Forest,” IJESPG (International J. Eng. Econ. Soc. Polit. Gov., vol. 1, no. 3, pp. 237–253, 2023.
[28] H. Blockeel, “Decision trees: from efficient prediction to responsible AI,” Front. Artif. Intell., 2023.
[29] R. A. Sembiring and Y. Sary, “Analisa dan Implementasi Long Short-Term Memory (LSTM) dalam Kebutuhan Persediaan Barang di PT. Gunung Sari Indonesia,” J. Ilm. Tek. Mesin, Elektro dan Komput., vol. 5, no. 3, pp. 97–105, 2025, doi: 10.51903/juritek.v5i3.5578.
[30] M. A. Saputra and T. Sugihartono, “Evaluasi Kinerja Model LSTM untuk Prediksi Risiko Penyakit Jantung Menggunakan Dataset,” J. Pendidik. dan Teknol. Indones., vol. 5, no. 7, pp. 1823–1833, 2025, doi: 10.52436/1.jpti.821.
[31] M. Tshamaroh, N. S. Nasution, N. Nadhirah, R. A. Alfira, and Z. Xintong, “Amazon Stock Price Prediction Using Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU),” Public Res. J. Eng. Data Technol. Comput. Sci., vol. 3, no. 1, pp. 1–9, 2025, doi: 10.57152/predatecs.v3i1.1656.
[32] Q. Yu, B. A. Tolson, H. Shen, M. Han, J. Mai, and J. Lin, “Enhancing Long Short-Term Memory (LSTM)-based Streamflow Prediction with a Spatially Distributed Approach,” Hydrol. Earth Syst. Sci., vol. 28, pp. 2107–2122, 2024, doi: 10.5194/hess-28-2107-2024.
[33] M. R. Humaidi and A. Maulani, “Klasifikasi Naïve Bayes dan Confusion Matrix pada Pengguna Aplikasi E-Commerce di Play Store,” J. Ilm. Inform., vol. 8, no. 2, pp. 132–139, 2023.
[34] N. Hidayah and D. Dodiman, “Implementasi algoritma multinomial naïve bayes, TF-IDF dan confusion matrix dalam pengklasifikasian saran monitoring dan evaluasi mahasiswa terhadap dosen teknik informatika Universitas Dayanu Ikhsanuddin,” in Jurnal Akademik Pendidikan Matematika, 2024, pp. 8–15.
[35] N. Nurdin, “Analisa Data Mining Dalam Memprediksi Masyarakat Kurang Mampu Menggunakan Metode K-Nearest Neighbor,” J. Inform. Dan Tek. Elektro Terap., vol. 12, no. 2, 2025.
[36] A. Botchkarev, “Performance Metrics (Error Measures) in Machine Learning Regression, Forecasting and Prognostics: Properties and Typology,” Interdiscip. J. Information, Knowledge, Manag., vol. 16, pp. 1–18, 2021.
[37] T. O. Hodson, “Root Mean Square Error (RMSE) or Mean Absolute Error (MAE): When to Use Which Metric,” Geosci. Model Dev., vol. 15, no. 14, pp. 5481–5493, 2022.
[38] C. C. Aggarwal, “Evaluation Metrics for Regression Models in Machine Learning,” ACM Comput. Surv., vol. 55, no. 6, pp. 1–36, 2023.
[39] A. Putra et al., “Prediction of Optical Fiber Network Attenuation to Assess Network Performance Using Random Forest Regression,” Int. J. Comput. Sci., vol. 20, no. 3, pp. 455–463, 2025.
[40] M. Soothar et al., “Hybrid CNN-LSTM Model for Fault Detection and Performance Analysis in Optical Fiber Networks,” Opt. Fiber Technol., vol. 78, pp. 103–112, 2024.
[41] Y. Zhang et al., “Prediction of Optical Fiber Cable Degradation Based on Bi-LSTM and Attention Mechanism,” Sensors, vol. 24, no. 12, p. 4012, 2024.
[42] R. Kumar et al., “LSTM-Based Model for Reliable Optical Transmission in Flexible Optical Networks,” Opt. Switch. Netw., vol. 47, pp. 100–109, 2023.
[43] X. Li et al., “AI-Driven Predictive Maintenance for DWDM Optical Fiber Networks Using Random Forest and LSTM,” Opt. Commun., vol. 540, pp. 129–138, 2025.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Hermansyah Hermansyah, Taufiq Taufiq, Defry Hamdhana, Munirul Ula, Muhammad Ikhwanus
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
