tuzsutТруды учебных заведений связиProceedings of Telecommunication Universities1813-324X2712-8830СПбГУТ10.31854/1813-324X-2023-9-5-66-78tuzsut-515Research ArticleИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИ И ТЕЛЕКОММУНИКАЦИИINFORMATION TECHNOLOGIES AND TELECOMMUNICATIONМодель классификации трафика в программно-конфигурируемых сетях c элементами искусственного интеллектаTraffic Classification Model in Software-Defined Networks with Artificial Intelligence Elementshttps://orcid.org/0000-0003-4213-953XЕлагинВ. С.ElaginV.

кандидат технических наук, доцент, доцент кафедры Инфокоммуникационных систем Санкт-Петербургского государственного университета телекоммуникаций им. проф. М.А. Бонч-Бруевича

v.elagin@sut.ruСанкт-Петербургский государственный университет телекоммуникаций им. М.А. Бонч-БруевичаРоссияThe Bonch-Bruevich Saint Petersburg State University of TelecommunicationsRussian Federation202314112023956678Copyright © Елагин В.С., 20232023Елагин В.С.Elagin V.This work is licensed under a Creative Commons Attribution 4.0 License.https://tuzs.sut.ru/jour/article/view/515

Классификация приложений необходима для повышения производительности сети. Однако при постоянном росте числа пользователей и приложений, а также масштабирования сетей, традиционные методы классификации не могут справляться в полной мере с идентификацией и классификацией сетевых приложений с необходимым уровнем задержки. Применение технологии глубокого обучения совместно с особенностями архитектуры программно-конфигурируемых сетей (SDN, аббр. от англ. Software-Defined Networking) позволит реализовать новую гибридную глубокую нейронную сеть для классификации приложений, которая сможет обеспечить высокую точность классификации без ручного выбора и извлечения признаков. В предлагаемой структуре предложена классификация приложений, с учетом логического централизованного управления на контроллере SDN. Обработанные данные используются для обучения гибридной глубокой нейронной сети, состоящей из многоуровневого автокодировщика, с высокой размерностью скрытого слоя и выходного слоя на базе регрессии softmax. Необходимые параметры сетевого потока могут быть получены при обработке трафика многоуровневым автокодировщиком вместо ручной обработки. Слой регрессии softmax используется в качестве конечного классификатора приложений. В статье приведены результаты моделирования, которые демонстрируют преимущества предложенного метода классификации, по сравнении с методом опорных векторов.

Application classification is essential to improve network performance. However, with the constant growth in the number of users and applications, as well as the scaling of networks, traditional classification methods cannot fully cope with the identification and classification of network applications with the required level of delay. The use of deep learning technology together with the architecture features of software-defined networks (SDN) will allow the implementation of a new hybrid deep neural network for application classification, which can provide high classification accuracy without manual selection and feature extraction. The proposed structure proposes a classification of applications, taking into account the logical centralized management on the SDN controller. The processed data is used to train a hybrid deep neural network consisting of stacked autoencoder with a high dimensionality of the hidden layer and an output layer based on softmax regression. The necessary network flow parameters can be obtained by processing traffic with a stacked auto-encoder instead of manual processing. The softmax regression layer is used as the final application classifier. The article presents simulation results that demonstrate the advantages of the proposed classification method in comparison with the support vector machine.

программно-конфигурируемые сетиПКСнейронная сетьклассификация трафикарегрессия softmaxsoftware-defined networksSDNneural networktraffic classificationsoftmax regressionнаучная статья подготовлена в рамках прикладных научный исследований СПбГУТ, регистрационный номер 123060900012-6 в ЕГИСУ НИОКТРthe article was prepared within the framework of applied scientific research of SPbSUT, registration number 123060900012-6 in the EGISU R&DReferencesЕлагин В.С. Динамическое управление нагрузкой в программно-конфигурируемых сетях // Труды учебных заведений связи. 2017. Т. 3. № 3. С. 60‒67.Elagin V. Dynamic Load Balancing in Software-Defined Network. Proceedings of Telecommun. Univ. 2017;3(3):60–67.Елагин В.С., Дмитриева Ю.С. Моделирование сетевого ресурса в программно-конфигурируемых сетях // Вестник связи. 2020. № 6. С. 35‒40.Elagin V.S., Dmitrieva Yu.S. The Modeling of Network Resources in Software-Defined Networks. Vestnik Communications. 2020;6:35‒40.Zhang J., Chen X., Xiang Y., Zhou W., Wu J. Robust Network Traffic Classification // IEEE /ACM Transactions on Networking. 2015. Vol. 23. Iss. 4. PP. 1257‒1270. DOI:10.1109/TNET.2014.2320577Zhang J., Chen X., Xiang Y., Zhou W., Wu J. Robust Network Traffic Classification. IEEE /ACM Transactions on Networking. 2015;23(4):1257‒1270. DOI:10.1109/TNET.2014.2320577Kim H., Claffy K.C., Fomenkov M., Barman D., Faloutsos M., Lee K. Internet traffic classification demystified: myths, caveats, and the best practices // Proceedings of the Conference on emerging Networking EXperiments and Technologies (Madrid, Spain, 9‒12 December 2008). New York: Association for Computing Machinery, 2008. DOI:10.1145/1544012.1544023Kim H., Claffy K.C., Fomenkov M., Barman D., Faloutsos M., Lee K. Internet traffic classification demystified: myths, caveats, and the best practices. Proceedings of the Conference on emerging Networking EXperiments and Technologies, 9‒12 December 2008, Madrid, Spain. New York: Association for Computing Machinery; 2008. DOI:10.1145/1544012.1544023Auld T., Moore A.W., Gull S.F. Bayesian Neural Networks for Internet Traffic Classification // IEEE Transactions Neural Networ. 2007. Vol. 18. Iss. 1. PP. 223‒239. DOI:10.1109/TNN.2006.883010Auld T., Moore A.W., Gull S.F. Bayesian Neural Networks for Internet Traffic Classification. IEEE Transactions Neural Networ. 2007;18(1):223‒239. DOI:10.1109/TNN.2006.883010Nguyen T.T.T., Armitage G. A survey of techniques for internet traffic classification using machine learning // IEEE Communication Survive Tutorials. 2008. Vol. 10. Iss. 4. PP. 56‒76. DOI:10.1109/SURV.2008.080406Nguyen T.T.T., Armitage G. A survey of techniques for internet traffic classification using machine learning. IEEE Communication Survive Tutorials. 2008;10(4):56‒76. DOI:10.1109/SURV.2008.080406Valenti S., Rossi D., Dainotti A., Pescapè A., Finamore A., Mellia M. Reviewing Traffic Classification // Biersack E., Callegari C., Matijasevic M. (eds) Data Traffic Monitoring and Analysis. Lecture Notes in Computer Science. Berlin, Germany: Springer, 2013. Vol. 7754. PP. 123‒147. DOI:10.1007/978-3-642-36784-7_6Valenti S., Rossi D., Dainotti A., Pescapè A., Finamore A., Mellia M. Reviewing Traffic Classification. In: Biersack E., Callegari C., Matijasevic M. (eds) Data Traffic Monitoring and Analysis. Lecture Notes in Computer Science, vol.7754. Berlin, Germany: Springer; 2013. p.123‒147. DOI:10.1007/978-3-642-36784-7_6Zhang J., Chen C., Xiang Y., Zhou W., Xiang Y. Internet Traffic Classification by Aggregating Correlated Naive Bayes Predictions // IEEE Transactions on Information Forensics and Security. 2013. Vol. 8. Iss. 1. PP. 5‒15. DOI:10.1109/TIFS.2012.2223675Zhang J., Chen C., Xiang Y., Zhou W., Xiang Y. Internet Traffic Classification by Aggregating Correlated Naive Bayes Predictions. IEEE Transactions on Information Forensics and Security. 2013:8(1):5‒15. DOI:10.1109/TIFS.2012.2223675Grimaudo L., Mellia M., Baralis E., Keralapura R. SeLeCT: Self-Learning Classifier for Internet Traffic // IEEE Transactions Network Service Management. 2014. Vol. 11. Iss. 2. PP. 144‒157. DOI:10.1109/TNSM.2014.011714.130505Grimaudo L., Mellia M., Baralis E., Keralapura R. SeLeCT: Self-Learning Classifier for Internet Traffic. IEEE Transactions Network Service Management. 2014;11(2):144‒157. DOI:10.1109/TNSM.2014.011714.130505Cao J., Fang Z., Qu G., Sun H., Zhang D. An accurate traffic classification model based on support vector machines // International Journal of Network Management. 2017. Vol. 27. Iss. 1. P. e1962. DOI:10.1002/nem.1962Cao J., Fang Z., Qu G., Sun H., Zhang D. An accurate traffic classification model based on support vector machines. International Journal of Network Management. 2017;27(1):e1962. DOI:10.1002/nem.1962Pasca S.T.V., Prasad S.S., Kataoka K. AMPF: Application-aware Multipath Packet Forwarding using Machine Learning and SDN // arXiv:1606.05743. 2016. DOI:10.48550/arXiv.1606.05743Pasca S.T.V., Prasad S.S., Kataoka K. AMPF: Application-aware Multipath Packet Forwarding using Machine Learning and SDN. arXiv:1606.05743. 2016. DOI:10.48550/arXiv.1606.05743Amaral P., Dinis J., Pinto P., Bernardo L., Tavares J., Mamede H.S. Machine Learning in Software Defined Networks: Data Collection and Traffic Classification // Proceedings of the 24th International Conference on Network Protocols (ICNP, Singapore, 08‒11 November 2016). IEEE, 2016. DOI:10.1109/ICNP.2016.7785327Amaral P., Dinis J., Pinto P., Bernardo L., Tavares J., Mamede H.S. Machine Learning in Software Defined Networks: Data Collection and Traffic Classification. Proceedings of the 24th International Conference on Network Protocols, ICNP, 08‒11 November 2016, Singapore. IEEE; 2016. DOI:10.1109/ICNP.2016.7785327Wang P., Lin S.C., Luo M. A Framework for QoS-aware Traffic Classification Using Semi-supervised Machine Learning in SDNs // Proceedings of the International Conference on Services Computing (SCC, San Francisco, USA, 27 June ‒ 02 July 2016). IEEE, 2016. DOI:10.1109/SCC.2016.133Wang P., Lin S.C., Luo M. A Framework for QoS-aware Traffic Classification Using Semi-supervised Machine Learning in SDNs. Proceedings of the International Conference on Services Computing, SCC, 27 June ‒ 02 July 2016, San Francisco, USA. IEEE; 2016. DOI:10.1109/SCC.2016.133LeCun Y., Bengio Y., Hinton G. Deep learning // Nature. 2015. Vol. 521. Iss. 7553. PP. 436‒444. DOI:10.1038/nature14539LeCun Y., Bengio Y., Hinton G. Deep learning. Nature. 2015;521(7553):436‒444. DOI:10.1038/nature14539Chen X.W., Lin X. Big Data Deep Learning: Challenges and Perspectives // IEEE Access. 2014. Vol. 2. PP. 514‒525. DOI:10.1109/ACCESS.2014.2325029Chen X.W., Lin X. Big Data Deep Learning: Challenges and Perspectives. IEEE Access. 2014;2:514‒525. DOI:10.1109/ACCESS.2014.2325029Kreutz D., Ramos F.M.V., Verissimo P.E., Rothenberg C.E., Azodolmolky S., Uhlig S. Software-Defined Networking: a Comprehensive Survey // Proceedings of the IEEE. 2015. Vol. 103. Iss. 1. PP. 14‒76. DOI:10.1109/JPROC.2014.2371999Kreutz D., Ramos F.M.V., Verissimo P.E., Rothenberg C.E., Azodolmolky S., Uhlig S. Software-Defined Networking: a Comprehensive Survey. Proceedings of the IEEE. 2015;103(1):14‒76. DOI:10.1109/JPROC.2014.2371999Bu C., Wang X., Cheng H., Huang M., Li K., Das S. Enabling Adaptive Routing Service Customization via the Integration of SDN and NFV // Journal of Network Computing Applications. 2017. Vol. 93. PP. 123‒136. DOI:10.1016/j.jnca.2017.05.010Bu C., Wang X., Cheng H., Huang M., Li K., Das S. Enabling Adaptive Routing Service Customization via the Integration of SDN and NFV. Journal of Network Computing Applications. 2017;93:123‒136. DOI:10.1016/j.jnca.2017.05.010Yi B., Wang X., Huang M. Design and evaluation of schemes for provisioning service function chainwith function scalability // Journal of Network Computing Applications. 2017. Vol. 93. PP. 197‒214. DOI:10.1016/j.jnca.2017.05.013Yi B., Wang X., Huang M. Design and evaluation of schemes for provisioning service function chainwith function scalability. Journal of Network Computing Applications. 2017;93:197‒214. DOI:10.1016/j.jnca.2017.05.013Lv J., Wang X., Huang M., Shi J., Li K., Li J. RISC: ICN routing mechanism incorporating SDN and community division // Computing Network. 2017. Vol. 123. PP. 88‒103. DOI:10.1016/j.comnet.2017.05.010Lv J., Wang X., Huang M., Shi J., Li K., Li J. RISC: ICN routing mechanism incorporating SDN and community division. Computing Network. 2017;123:88‒103. DOI:10.1016/j.comnet.2017.05.010He Q., Wang X., Huang M. OpenFlow-based low-overhead and high-accuracy SDN measurement framework // Transactions on Emerging Telecommunications Technologies. 2018. Vol. 29. Iss. 2. P. e3263. DOI:10.1002/ett.3263He Q., Wang X., Huang M. OpenFlow-based low-overhead and high-accuracy SDN measurement framework. Transactions on Emerging Telecommunications Technologies. 2018;29(2):e3263. DOI:10.1002/ett.3263Yi B., Wang X., Li K., Das S.K., Huang M. A comprehensive survey of Network Function Virtualization // Computing Network. 2018. Vol. 133. PP. 212‒262. DOI:10.1016/j.comnet.2018.01.021Yi B., Wang X., Li K., Das S.K., Huang M. A comprehensive survey of Network Function Virtualization. Computing Network. 2018;133:212‒262. DOI:10.1016/j.comnet.2018.01.021Shu Z., Wan J., Lin J., Wang S., Li D., Rho S., et al. Traffic engineering in software-defined networking: Measurement and management // IEEE Access. 2016. Vol. 4. PP. 3246‒3256. DOI:10.1109/ACCESS.2016.2582748Shu Z., Wan J., Lin J., Wang S., Li D., Rho S., et al. Traffic engineering in software-defined networking: Measurement and management. IEEE Access. 2016;4:3246‒3256. DOI:10.1109/ACCESS.2016.2582748Cui L., Yu F.R., Yan Q. When big data meets software-defined networking: SDN for big data and big data for SDN // IEEE Network. 2016. Vol. 30. Iss. 1. PP. 58‒65. DOI:10.1109/MNET.2016.7389832Cui L., Yu F.R., Yan Q. When big data meets software-defined networking: SDN for big data and big data for SDN. IEEE Network. 2016;30(1):58‒65. DOI:10.1109/MNET.2016.7389832Zhang L., Huang H., Jing X. A modified cyclostationary spectrum sensing based on softmax regression model // Proceedings of the 16th International Symposium on Communications and Information Technologies (ISCIT, Qingdao, China, 26‒28 September 2016). IEEE, 2016. DOI:10.1109/ISCIT.2016.7751707Zhang L., Huang H., Jing X. A modified cyclostationary spectrum sensing based on softmax regression model. Proceedings of the 16th International Symposium on Communications and Information Technologies, ISCIT, 26‒28 September 2016, Qingdao, China. IEEE; 2016. DOI:10.1109/ISCIT.2016.7751707Zhang H., Lu G., Qassrawi M.T., Zhang Y., Yu X. Feature selection for optimizing traffic classification // Computing Communicdtion. 2012. Vol. 35. Iss. 12. PP. 1457‒1471. DOI:10.1016/j.comcom.2012.04.012Zhang H., Lu G., Qassrawi M.T., Zhang Y., Yu X. Feature selection for optimizing traffic classification. Computing Communicdtion. 2012;35(12):1457‒1471. DOI:10.1016/j.comcom.2012.04.012da Silva A.S., Machado C.C., Bisol R.V., Granville L.Z., Schaeffer A. Identification and Selection of Flow Features for Accurate Traffic Classification in SDN // Proceedings of the 14th International Symposium on Network Computing and Applications (NCA, Cambridge, USA, 28‒30 September 2015). IEEE, 2015. DOI:10.1109/NCA.2015.12da Silva A.S., Machado C.C., Bisol R.V., Granville L.Z., Schaeffer A. Identification and Selection of Flow Features for Accurate Traffic Classification in SDN. Proceedings of the 14th International Symposium on Network Computing and Applications, USA, 28‒30 September 2015, NCA, Cambridge. IEEE; 2015. DOI:10.1109/NCA.2015.12Schmidhuber J. Deep learning in neural networks: an overview // Neural Network. 2015. Vol. 61. PP. 85‒117. DOI:10.1016/j.neunet.2014.09.003Schmidhuber J. Deep learning in neural networks: an overview. Neural Network. 2015;61:85‒117. DOI:10.1016/j.neunet.2014.09.003Salama M.A., Eid H.F., Ramadan R.A., Darwish A., Hassanien E. Hybrid Intelligent Intrusion Detection Scheme // Gaspar-Cunha A., Takahashi R., Schaefer G., Costa L. (eds) Soft Computing in Industrial Applications. Advances in Intelligent and Soft Computing. Berlin, Heidelberg: Springer, 2011. Vol. 96. PP. 293‒303. DOI:10.1007/978-3-642-20505-7_26Salama M.A., Eid H.F., Ramadan R.A., Darwish A., Hassanien E. Hybrid Intelligent Intrusion Detection Scheme. In: Gaspar-Cunha A., Takahashi R., Schaefer G., Costa L. (eds) Soft Computing in Industrial Applications. Advances in Intelligent and Soft Computing, vol.96. Berlin, Heidelberg: Springer; 2011. p.293‒303. DOI:10.1007/978-3-642-20505-7_26Fiore U., Palmieri F., Castiglione A., De Santis A. Network anomaly detection with the restricted Boltzmann machine // Neurocomputing. 2013. Vol. 122. PP. 13‒23. DOI:10.1016/j.neucom.2012.11.050Fiore U., Palmieri F., Castiglione A., De Santis A. Network anomaly detection with the restricted Boltzmann machine. Neurocomputing. 2013;122:13‒23. DOI:10.1016/j.neucom.2012.11.050Lv Y., Duan Y., Kang W., Li Z., Wang F.Y. Traffic Flow Prediction with Big Data: a Deep Learning Approach // IEEE Transactions Intelligent Transport System. 2015. Vol. 16. Iss. 2. PP. 865‒873. DOI:10.1109/TITS.2014.2345663Lv Y., Duan Y., Kang W., Li Z., Wang F.Y. Traffic Flow Prediction with Big Data: a Deep Learning Approach. IEEE Transactions Intelligent Transport System. 2015;16(2):865‒873. DOI:10.1109/TITS.2014.2345663Yang H.F., Dillon T.S., Chen Y.P. Optimized Structure of the Traffic Flow Forecasting Model with a Deep Learning Approach // IEEE Transactions on Neural Networks and Learning Systems. 2017. Vol. 28. Iss. 10. PP. 2371‒2381. DOI:10.1109/TNNLS.2016.2574840Yang H.F., Dillon T.S., Chen Y.P. Optimized Structure of the Traffic Flow Forecasting Model with a Deep Learning Approach. IEEE Transactions on Neural Networks and Learning Systems. 2017;28(10):2371‒2381. DOI:10.1109/TNNLS.2016.2574840Huang W., Song G., Hong H., Xie K. Deep Architecture for Traffic Flow Prediction: Deep Belief Networks with Multitask Learning // IEEE Transactions Intelligent Transport System. 2014. Vol. 15. Iss. 5. PP. 2191‒2201. DOI:10.1109/TITS.2014.2311123Huang W., Song G., Hong H., Xie K. Deep Architecture for Traffic Flow Prediction: Deep Belief Networks with Multitask Learning. IEEE Transactions Intelligent Transport System. 2014;15(5):2191‒2201. DOI:10.1109/TITS.2014.2311123Bengio Y., Lamblin P., Popovici D., Larochelle H. Greedy Layer-Wise Training of Deep Networks // Proceedings of the Conference on Advances in Neural Information Processing Systems 19 (2006). MIT Press, 2007. PP. 153‒160.Bengio Y., Lamblin P., Popovici D., Larochelle H. Greedy Layer-Wise Training of Deep Networks. Proceedings of the Conference on Advances in Neural Information Processing Systems 19, 2006. MIT Press; 2007. p.153‒160.BRASIL. Characterizing Network-based Applications. Data sets // University of Cambridge Computer Laboratory. URL: https://www.cl.cam.ac.uk/research/srg/netos/projects/brasil/data/index.html (дата обращения 15.06.2023)University of Cambridge Computer Laboratory. BRASIL. Characterizing Network-based Applications. Data sets. URL: https://www.cl.cam.ac.uk/research/srg/netos/projects/brasil/data/index.html [Accessed 15.06.2023]

The authors declare that there are no conflicts of interest present.