From Statistics to Deep Learning: Forecasting Mobile Throughput
Abstract
Accurate download throughput prediction is critical for adaptive resource management and QoS in 5G networks, particularly under high user mobility. This work systematically investigates two key design choices in time series forecasting: (i) local versus global models and (ii) the inclusion or exclusion of external covariates. We evaluate statistical, machine learning, and deep learning methods on real-world 5G data, where throughput is predicted using channel quality metrics and user speed as potential covariates. Experimental results show that global, tree-based ensembles like LightGBM achieve the best trade-off between accuracy, robustness, and efficiency. Furthermore, we found that the explored network quality covariates were insufficient to consistently improve performance for this complex task. All source code is available at: https://github.com/ejs94/5g-forecasting.
Keywords:
time series forecasting, 5G networks, throughput prediction, machine learning, mobility, quality of service
References
Batool, I., Fouda, M. M., and Fadlullah, Z. M. Deep Learning-Based Throughput Prediction in 5G Cellular Networks. In 2024 International Conference on Smart Applications, Communications and Networking (SmartNets). Institute of Electrical and Electronics Engineers (IEEE), 2024.
Boutiba, K., Bagaa, M., and Ksentini, A. Radio link failure prediction in 5G networks. In 2021 IEEE Global Communications Conference (GLOBECOM). IEEE, pp. 1–6, 2021.
Elsherbiny, H., Abbas, H. M., Abou-zeid, H., Hassanein, H. S., and Noureldin, A. 4g lte network throughput modelling and prediction. In GLOBECOM 2020 - 2020 IEEE Global Communications Conference. IEEE, pp. 1–6, 2020.
Ghosh, A., Maeder, A., Baker, M., and Chandramouli, D. 5G Evolution: A View on 5G Cellular Technology beyond 3GPP Release 15. IEEE Access vol. 7, pp. 127639–127651, 2019.
Hewamalage, H., Ackermann, K., and Bergmeir, C. Forecast evaluation for data scientists: common pitfalls and best practices. Data Mining and Knowledge Discovery vol. 37, pp. 788–832, 3, 2023.
Hyndman, R. J. A brief history of forecasting competitions. International Journal of Forecasting vol. 36, pp. 7–14, 1, 2020.
Hyndman, R. J. and Athanasopoulos, G. Forecasting: Principles and Practice. OTexts, Melbourne, Australia, 2021.
Januschowski, T., Gasthaus, J., Wang, Y., Salinas, D., Flunkert, V., Bohlke-Schneider, M., and Callot, L. Criteria for classifying forecasting methods. International Journal of Forecasting vol. 36, pp. 167–177, 1, 2020.
Makridakis, S., Spiliotis, E., and Assimakopoulos, V. M5 accuracy competition: Results, findings, and conclusions. International Journal of Forecasting vol. 38, pp. 1346–1364, 10, 2022.
Narayanan, A., Ramadan, E., Mehta, R., Hu, X., Liu, Q., Fezeu, R. A. K., Dayalan, U. K., Verma, S., Ji, P., Li, T., Qian, F., and Zhang, Z.-L. Lumos5G: Mapping and Predicting Commercial mmWave 5G Throughput. In Proceedings of the ACM SIGCOMM Internet Measurement Conference (IMC ’20). Association for Computing Machinery (ACM), pp. 176–193, 2020.
Raca, D., Leahy, D., Sreenan, C. J., and Quinlan, J. J. Beyond throughput, the next generation: A 5G dataset with channel and context metrics. MMSys 2020 - Proceedings of the 2020 Multimedia Systems Conference, 5, 2020.
Santos, G. L., Endo, P. T., Sadok, D., and Kelner, J. When 5g meets deep learning: A systematic review. Algorithms vol. 13, pp. 208, aug, 2020.
Sharma, A., Pandit, S., and Talluri, S. R. Throughput prediction of fifth-generation cellular system using hybrid feature selection and enhanced sequential decision tree machine learning algorithm. Wireless Networks vol. 31, pp. 3025–3042, 2025.
Yeaser, K. M. A. and Hassan, K. M. A. 5G NR V2X Throughput Prediction Using Deep Hybrid Learning. In Innovations in Electrical and Electronics Engineering: Proceedings of the 5th ICIEEL 2024, A. Kalam, S. Mekhilef, and S. S. Williamson (Eds.). Springer Nature Singapore, Singapore, pp. 685–693, 2025.
Yingjie, Z. and Abolghasemi, M. Local vs. global models for hierarchical forecasting, 2024. Available at: [link]. Accessed on: 5 Aug. 2025.
Boutiba, K., Bagaa, M., and Ksentini, A. Radio link failure prediction in 5G networks. In 2021 IEEE Global Communications Conference (GLOBECOM). IEEE, pp. 1–6, 2021.
Elsherbiny, H., Abbas, H. M., Abou-zeid, H., Hassanein, H. S., and Noureldin, A. 4g lte network throughput modelling and prediction. In GLOBECOM 2020 - 2020 IEEE Global Communications Conference. IEEE, pp. 1–6, 2020.
Ghosh, A., Maeder, A., Baker, M., and Chandramouli, D. 5G Evolution: A View on 5G Cellular Technology beyond 3GPP Release 15. IEEE Access vol. 7, pp. 127639–127651, 2019.
Hewamalage, H., Ackermann, K., and Bergmeir, C. Forecast evaluation for data scientists: common pitfalls and best practices. Data Mining and Knowledge Discovery vol. 37, pp. 788–832, 3, 2023.
Hyndman, R. J. A brief history of forecasting competitions. International Journal of Forecasting vol. 36, pp. 7–14, 1, 2020.
Hyndman, R. J. and Athanasopoulos, G. Forecasting: Principles and Practice. OTexts, Melbourne, Australia, 2021.
Januschowski, T., Gasthaus, J., Wang, Y., Salinas, D., Flunkert, V., Bohlke-Schneider, M., and Callot, L. Criteria for classifying forecasting methods. International Journal of Forecasting vol. 36, pp. 167–177, 1, 2020.
Makridakis, S., Spiliotis, E., and Assimakopoulos, V. M5 accuracy competition: Results, findings, and conclusions. International Journal of Forecasting vol. 38, pp. 1346–1364, 10, 2022.
Narayanan, A., Ramadan, E., Mehta, R., Hu, X., Liu, Q., Fezeu, R. A. K., Dayalan, U. K., Verma, S., Ji, P., Li, T., Qian, F., and Zhang, Z.-L. Lumos5G: Mapping and Predicting Commercial mmWave 5G Throughput. In Proceedings of the ACM SIGCOMM Internet Measurement Conference (IMC ’20). Association for Computing Machinery (ACM), pp. 176–193, 2020.
Raca, D., Leahy, D., Sreenan, C. J., and Quinlan, J. J. Beyond throughput, the next generation: A 5G dataset with channel and context metrics. MMSys 2020 - Proceedings of the 2020 Multimedia Systems Conference, 5, 2020.
Santos, G. L., Endo, P. T., Sadok, D., and Kelner, J. When 5g meets deep learning: A systematic review. Algorithms vol. 13, pp. 208, aug, 2020.
Sharma, A., Pandit, S., and Talluri, S. R. Throughput prediction of fifth-generation cellular system using hybrid feature selection and enhanced sequential decision tree machine learning algorithm. Wireless Networks vol. 31, pp. 3025–3042, 2025.
Yeaser, K. M. A. and Hassan, K. M. A. 5G NR V2X Throughput Prediction Using Deep Hybrid Learning. In Innovations in Electrical and Electronics Engineering: Proceedings of the 5th ICIEEL 2024, A. Kalam, S. Mekhilef, and S. S. Williamson (Eds.). Springer Nature Singapore, Singapore, pp. 685–693, 2025.
Yingjie, Z. and Abolghasemi, M. Local vs. global models for hierarchical forecasting, 2024. Available at: [link]. Accessed on: 5 Aug. 2025.
Published
2025-09-29
How to Cite
SANTOS, Estêvão J.; ALMEIDA, Tiago A..
From Statistics to Deep Learning: Forecasting Mobile Throughput. In: SYMPOSIUM ON KNOWLEDGE DISCOVERY, MINING AND LEARNING (KDMILE), 13. , 2025, Fortaleza/CE.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 137-144.
ISSN 2763-8944.
DOI: https://doi.org/10.5753/kdmile.2025.247767.
