Inferência da qualidade do serviço em enlaces de rede através de um método baseado em aprendizado federado
Resumo
A crescente demanda de acesso à Internet acarreta no aumento da demanda por melhorias nas infraestruturas das redes provedoras de Internet e redes de pontos de acesso móvel, e a complexidade deste novo cenário motiva o uso de técnicas de aprendizado de máquina para monitoramento e gerenciamento de redes. Este trabalho propõe um método baseado em aprendizado federado para inferir a qualidade de serviço na rede. Como entrada deste modelo, são usadas matrizes de tráfego contendo medidas coletadas nos próprios nós de origem. Utilizando criptografia homomórfica, os nós de origem mantêm a privacidade dos seus dados de medição, prevenindo-os de ataques de vazamento de informação durante o processo de treinamento. Resultados dos experimentos realizados apontam que o método proposto superou o aprendizado de máquina centralizado em termos de desempenho, e com um menor número de rodadas até a convergência do modelo.
Referências
Acar, A., Aksu, H., Uluagac, A. S., and Conti, M. (2018). A survey on homomorphic encryption schemes: Theory and implementation. ACM Computing Surveys, 51(4):1–35.
Amoroso, R., Esposito, F., and Merani, M. L. (2020). Estimation of traffic matrices via super-resolution and federated learning. In 16th ACM CoNEXT, pages 560–561.
Aono, Y., Hayashi, T., Wang, L., Moriai, S., et al. (2017). Privacy-preserving deep learning via additively homomorphic encryption. IEEE Transactions on Information Forensics and Security, 13(5):1333–1345.
Bishop, C. M. and Nasrabadi, N. M. (2006). Pattern recognition and machine learning, volume 4. Springer.
Clausen, H., Manocha, A., and Gibson, M. (2022). Detecting proxies relaying streaming internet traffic. In IEEE INFOCOM WKSHPS, pages 1–6.
Fawcett, T. (2006). An introduction to roc analysis. Pattern recognition letters, 27(8):861–874.
Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep learning. MIT press.
Grigoriou, E., Saoulidis, T., Atzori, L., Pilloni, V., and Chatzimisios, P. (2018). An agent-based qoe monitoring strategy for lte networks. In IEEE ICC, pages 1–6.
Hartmann, F. (2018). Federated learning. Master’s thesis, Institut fur Informatik der Freien Universität Berlin, Available on: https://florian.github.io/federated-learning/.
Lo, S. K., Lu, Q., Wang, C., Paik, H.-Y., and Zhu, L. (2021). A systematic literature review on federated machine learning: From a software engineering perspective. ACM Computing Surveys, 54(5):1–39.
McMahan, B., Moore, E., Ramage, D., Hampson, S., and Arcas, B. A. y. (2017). Communication-Efficient Learning of Deep Networks from Decentralized Data. In 20th AISTATS, pages 1273–1282.
McMahan, H. B., Moore, E., Ramage, D., and y Arcas, B. A. (2016). Federated learning of deep networks using model averaging. arXiv preprint arXiv:1602.05629.
Mesquita, L. A. J. and Assis, K. D. R. (2019). Traffic matrix prediction for optical networks. In IEEE IMOC, pages 1–3.
Monteiro, N., Fontinele, A., Santos, I., Costa, A., Campelo, D., and Soares, A. (2019). Novo algoritmo para provisão de banda de guarda adaptativa em redes Ópticas elásticas. In XXXVII SBRC, pages 307–320.
Navarro-Ortiz, J., Romero-Diaz, P., Sendra, S., Ameigeiras, P., Ramos-Munoz, J. J., and Lopez-Soler, J. M. (2020). A survey on 5g usage scenarios and traffic models. IEEE Communications Surveys & Tutorials, 22(2):905–929.
She, H., Zhu, X., Guo, Y., Cao, H., Garg, S., and Kaddoum, G. (2022). Bclb: Blockchain-based controller load balance for safe and reliable resource optimization. In IEEE Globecom Workshops, pages 668–673.
Thonglek, K., Takahashi, K., Ichikawa, K., Iida, H., and Nakasan, C. (2020). Federated learning of neural network models with heterogeneous structures. In 19th IEEE ICMLA, pages 735–740.
Tu, R., Wang, X., Zhao, J., Yang, Y., Shi, L., and Wolf, T. (2015). Design of a load-balancing middlebox based on sdn for data centers. In IEEE INFOCOM WKSHPS, pages 480–485.
Uhlig, S., Quoitin, B., Lepropre, J., and Balon, S. (2006). Providing public intradomain traffic matrices to the research community. ACM SIGCOMM Computer Communication Review, 36(1):83–86.
Amoroso, R., Esposito, F., and Merani, M. L. (2020). Estimation of traffic matrices via super-resolution and federated learning. In 16th ACM CoNEXT, pages 560–561.
Aono, Y., Hayashi, T., Wang, L., Moriai, S., et al. (2017). Privacy-preserving deep learning via additively homomorphic encryption. IEEE Transactions on Information Forensics and Security, 13(5):1333–1345.
Bishop, C. M. and Nasrabadi, N. M. (2006). Pattern recognition and machine learning, volume 4. Springer.
Clausen, H., Manocha, A., and Gibson, M. (2022). Detecting proxies relaying streaming internet traffic. In IEEE INFOCOM WKSHPS, pages 1–6.
Fawcett, T. (2006). An introduction to roc analysis. Pattern recognition letters, 27(8):861–874.
Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep learning. MIT press.
Grigoriou, E., Saoulidis, T., Atzori, L., Pilloni, V., and Chatzimisios, P. (2018). An agent-based qoe monitoring strategy for lte networks. In IEEE ICC, pages 1–6.
Hartmann, F. (2018). Federated learning. Master’s thesis, Institut fur Informatik der Freien Universität Berlin, Available on: https://florian.github.io/federated-learning/.
Lo, S. K., Lu, Q., Wang, C., Paik, H.-Y., and Zhu, L. (2021). A systematic literature review on federated machine learning: From a software engineering perspective. ACM Computing Surveys, 54(5):1–39.
McMahan, B., Moore, E., Ramage, D., Hampson, S., and Arcas, B. A. y. (2017). Communication-Efficient Learning of Deep Networks from Decentralized Data. In 20th AISTATS, pages 1273–1282.
McMahan, H. B., Moore, E., Ramage, D., and y Arcas, B. A. (2016). Federated learning of deep networks using model averaging. arXiv preprint arXiv:1602.05629.
Mesquita, L. A. J. and Assis, K. D. R. (2019). Traffic matrix prediction for optical networks. In IEEE IMOC, pages 1–3.
Monteiro, N., Fontinele, A., Santos, I., Costa, A., Campelo, D., and Soares, A. (2019). Novo algoritmo para provisão de banda de guarda adaptativa em redes Ópticas elásticas. In XXXVII SBRC, pages 307–320.
Navarro-Ortiz, J., Romero-Diaz, P., Sendra, S., Ameigeiras, P., Ramos-Munoz, J. J., and Lopez-Soler, J. M. (2020). A survey on 5g usage scenarios and traffic models. IEEE Communications Surveys & Tutorials, 22(2):905–929.
She, H., Zhu, X., Guo, Y., Cao, H., Garg, S., and Kaddoum, G. (2022). Bclb: Blockchain-based controller load balance for safe and reliable resource optimization. In IEEE Globecom Workshops, pages 668–673.
Thonglek, K., Takahashi, K., Ichikawa, K., Iida, H., and Nakasan, C. (2020). Federated learning of neural network models with heterogeneous structures. In 19th IEEE ICMLA, pages 735–740.
Tu, R., Wang, X., Zhao, J., Yang, Y., Shi, L., and Wolf, T. (2015). Design of a load-balancing middlebox based on sdn for data centers. In IEEE INFOCOM WKSHPS, pages 480–485.
Uhlig, S., Quoitin, B., Lepropre, J., and Balon, S. (2006). Providing public intradomain traffic matrices to the research community. ACM SIGCOMM Computer Communication Review, 36(1):83–86.
Publicado
26/05/2023
Como Citar
SILVA, Nicolas A. Alves da; CAMPOS, Carlos A. Vieira; LUCENA, Sidney C. de.
Inferência da qualidade do serviço em enlaces de rede através de um método baseado em aprendizado federado. In: WORKSHOP DE GERÊNCIA E OPERAÇÃO DE REDES E SERVIÇOS (WGRS), 28. , 2023, Brasília/DF.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2023
.
p. 71-84.
ISSN 2595-2722.
DOI: https://doi.org/10.5753/wgrs.2023.767.
