Iterative Network Throughput Prediction Using a Sliding Window Approach Combined with Dynamic Data Imputation

• Maria C. Mesquita UECE
• Maria L. Linhares UECE
• Ariel L. Portela UECE
• Ivo A. Pimenta UECE
• Thelmo P. Araújo UECE
• Rafael L. Gomes UECE

DOI: https://doi.org/10.5753/sbrc.2026.19334

Abstract

Throughput prediction is essential for proactive network management in Internet Service Providers (ISPs); however, real-world time series often exhibit gaps resulting from limitations in data collection mechanisms. This paper proposes an approach that operates in an autoregressive manner, utilizing actual values when available and predictions when observed data are missing, thereby integrating fault handling directly into the predictive process. Experiments conducted with real-world data from Ipê network infraestructure from National Education and Research Network (RNP) demonstrate that the approach maintains consistent performance even in scenarios with significant measurement failures, establishing it as a viable solution for supporting preventive network management.

References

Ahn, H., Sun, K., and Kim, K. (2021). Comparison of missing data imputation methods in time series forecasting. Computers, Materials and Continua, 70:767–779.

Al-Thaedan, A., Shakir, Z., Mjhool, A. Y., Alsabah, R., Al-Sabbagh, A., Salah, M., and Zec, J. (2023). Downlink throughput prediction using machine learning models on 4g-lte networks. International Journal of Information Technology, 15(6):2987–2993.

Brito, M. L. L., Ferreira, M. C. M., Portela, A. L. C., and Gomes, R. L. (2026). Ai-based estimation of bandwidth availability for data offloading in edge-cloud computing. IEEE Networking Letters, 8:69–73.

Du, W., Wang, J., Qian, L., Yang, Y., Ibrahim, Z., Liu, F., Wang, Z., Liu, H., Zhao, Z., Zhou, Y., Wang, W., Ding, K., Liang, Y., Prakash, B. A., and Wen, Q. (2024). Tsi-bench: Benchmarking time series imputation.

Ferreira, M., Linhares, M., Araújo, T., and Gomes, R. (2025). Aplicando decomposição de valores singulares na predição de vazão de rede. In Anais do XLIII Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos, pages 630–643, Porto Alegre, RS, Brasil. SBC.

Ferreira, M. C., Ribeiro, S. E., Nobre, F. V., Linhares, M. L., Araújo, T. P., and Gomes, R. L. (2024). Mitigating measurement failures in throughput performance forecasting. In 2024 20th International Conference on Network and Service Management (CNSM), pages 1–7.

Gomes, R. L., Bittencourt, L. F., and Madeira, E. R. (2014). A bandwidth-feasibility algorithm for reliable virtual network allocation. In 2014 IEEE 28th International Conference on Advanced Information Networking and Applications, pages 504–511.

Gomes, R. L., Bittencourt, L. F., Madeira, E. R., Cerqueira, E., and Gerla, M. (2016). Bandwidth-aware allocation of resilient virtual software defined networks. Computer Networks, 100:179–194.

Gomes, R. L., Bittencourt, L. F., and Madeira, E. R. M. (2013). A framework for sla establishment of virtual networks based on qos classes. In 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013), pages 1175–1178.

Hu, H., Qian, S., Yang, D., Cao, J., and Xue, G. (2024). Iterative time series imputation by maintaining dependency consistency. ACM Trans. Knowl. Discov. Data, 19(1).

Kablaoui, R., Ahmad, I., Abed, S., and Awad, M. (2024). Network traffic prediction by learning time series as images. Engineering Science and Technology, an International Journal, 55:101754.

Li, X. (2024). Time series forecasting with missing data using generative adversarial networks and bayesian inference. Information, 15(4).

Lopes Gomes, R. and Roberto Mauro Madeira, E. (2012). A traffic classification agent for virtual networks based on qos classes. IEEE Latin America Transactions, 10(3):1734–1741.

Mutter, E. and Shannigrahi, S. (2024). Science dmz networks: How different are they really? In 2024 IEEE 49th Conference on Local Computer Networks (LCN), page 1–9. IEEE.

Na, H., Shin, Y., Lee, D., and Lee, J. (2023). Lstm-based throughput prediction for lte networks. ICT Express, 9(2):247–252.

Neog, A., Daw, A., Khorasgani, S. F., Sawhney, M., Pradhan, A., Lofton, M. E., McAfee, B. J., Breef-Pilz, A., Wander, H. L., Howard, D. W., Carey, C. C., Hanson, P., and Karpatne, A. (2026). Investigating a model-agnostic and imputation-free approach for irregularly-sampled multivariate time-series modeling.

Nobre, F. V. J., Silva, D. d. S., Ferreira, M. C. M. M., Brito, M. L. M. L., de Araújo, T. P., and Gomes, R. L. (2025). Time-weighted correlation approach to identify high delay links in internet service providers. Journal of Internet Services and Applications, 16(1):419–430.

Pimenta, I., Silva, D., Moura, E., Silveira, M., and Gomes, R. L. (2024). Impact of data anonymization in machine learning models. In Proceedings of the 13th Latin-American Symposium on Dependable and Secure Computing, pages 188–191.

Portela, A., Linhares, M. M., Nobre, F. V. J., Menezes, R., Mesquita, M., and Gomes, R. L. (2024). The role of tcp congestion control in the throughput forecasting. In Proceedings of the 13th Latin-American Symposium on Dependable and Secure Computing, pages 196–199.

Qian, L., Yang, Y., Du, W., Wang, J., Dobsoni, R., and Ibrahim, Z. (2025). Beyond random missingness: Clinically rethinking for healthcare time series imputation.

Yalda, K., Jamal Hamad, D., Tapus, N., and Okumus, I. T. (2024). Network traffic prediction performance using lstm. Romanian Journal of Information Science and Technology, 27:336–347.