Deep Learning Networks for Classification and Congestion Control in TCP/IP Networks

• Cesar Augusto C. Marcondes ITA
• Marcelo R. da Siva ITA

DOI: https://doi.org/10.5753/sbrc.2024.1253

Abstract

The advancement and ubiquity of digital networks have fundamentally transformed numerous spheres of human activity. At the heart of this phenomenon, lies the Transmission Control Protocol (TCP) model, whose influence is particularly notable in the exponential growth of the Internet due to its ability to transmit flexibly to any device, through its advanced Congestion Control (CC). Seeking an even more efficient CC mechanism, this work proposes the construction of deep learning neural networks (MLP, LSTM, and CNN) for classifying the level of network congestion. The results attest to models capable of distinguishing, with over 90% accuracy, between moments of high and low degrees of congestion. With this, it becomes possible to differentiate between congestion and random losses, potentially increasing throughput by up to five times in environments with random losses when combined with CC algorithms.

References

Abbasloo, S., Yen, C.-Y., and Chao, H. J. (2020). Classic Meets Modern: a Pragmatic Learning-Based Congestion Control for the Internet. In Proceedings of the Annual conference of the ACM Special Interest Group on Data Communication on the applications, technologies, architectures, and protocols for computer communication, pages 632–647, Virtual Event USA. ACM.

Badarla, V. and Murthy, C. S. R. (2011). Learning-TCP: A stochastic approach for efficient update in TCP congestion window in ad hoc wireless networks. Journal of Parallel and Distributed Computing, 71(6):863–878.

Bai, L., Abe, H., and Lee, C. (2020). RNN-based Approach to TCP throughput prediction. In 2020 Eighth International Symposium on Computing and Networking Workshops (CANDARW), pages 391–395, Naha, Japan. IEEE.

Brakmo, L. S., O’Malley, S. W., and Peterson, L. L. (1994). Tcp vegas: New techniques for congestion detection and avoidance. In Proceedings of the Conference on Communications Architectures, Protocols and Applications, SIGCOMM ’94, page 24–35, New York, NY, USA. Association for Computing Machinery.

Conlin, R., Erickson, K., Abbate, J., and Kolemen, E. (2021). Keras2c: A library for converting Keras neural networks to real-time compatible C. Engineering Applications of Artificial Intelligence, 100:104182. 0000063.

Emara, S., Li, B., and Chen, Y. (2020). Eagle: Refining Congestion Control by Learning from the Experts. In IEEE INFOCOM 2020 - IEEE Conference on Computer Communications, pages 676–685, Toronto, ON, Canada. IEEE.

Fukushima, K. (1980). Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 36(4):193–202. 0000061.

Hochreiter, S. and Schmidhuber, J. (1997). Long Short-Term Memory. Neural Computation, 9(8):1735–1780. 0000060 Conference Name: Neural Computation.

ITU, G. C. R. (2023). Global Connectivity Report 2022 - [link] acesso em 13/12/23.

Jay, N., Rotman, N., Godfrey, B., Schapira, M., and Tamar, A. (2019). A deep reinforcement learning perspective on internet congestion control. In Chaudhuri, K. and Salakhutdinov, R., editors, Proceedings of the 36th International Conference on Machine Learning, volume 97 of Proceedings of Machine Learning Research, pages 3050–3059. PMLR.

Kazama, R., Abe, H., and Lee, C. (2022). Evaluating TCP throughput predictability from packet traces using recurrent neural network. In 2022 IEEE Symposium on Computers and Communications (ISCC), pages 1–6. 0000057 ISSN: 2642-7389.

Li, W., Zhou, F., Chowdhury, K. R., and Meleis, W. (2019). QTCP: Adaptive Congestion Control with Reinforcement Learning. IEEE Transactions on Network Science and Engineering, 6(3):445–458.

Li, W., Zhou, F., Meleis, W., and Chowdhury, K. (2016). Learning-Based and Data-Driven TCP Design for Memory-Constrained IoT. In 2016 International Conference on Distributed Computing in Sensor Systems (DCOSS), pages 199–205, Washington, DC, USA. IEEE.

Lippmann, R. (1987). An introduction to computing with neural nets. IEEE ASSP Magazine, 4(2):4–22. 0000062 Conference Name: IEEE ASSP Magazine.

Lu, J., Liu, A., Dong, F., Gu, F., Gama, J., and Zhang, G. (2018). Learning under Concept Drift: A Review. IEEE Transactions on Knowledge and Data Engineering, pages 1–1. 0000065 arXiv:2004.05785 [cs, stat].

Ng, S. W. and Chan, E. (2005). Equation-based TCP-friendly congestion control under lossy environment. Journal of Systems Architecture, 51(9):542–569.

ns3 (2023). A discrete-event network simulator for internet systems - [link] - vistado em 13/12/05.

Ramana, B., Manoj, B., and Murthy, C. (2005). Learning-TCP: a novel learning automata based reliable transport protocol for ad hoc wireless networks. In 2nd International Conference on Broadband Networks, 2005., pages 521–530, Boston, MA. IEEE.

Sutton, R. S. and Barto, A. G. (2018). Reinforcement Learning: An Introduction. The MIT Press, second edition.

Winstein, K. and Balakrishnan, H. (2013). Tcp ex machina: Computer-generated congestion control. In Proceedings of the ACM SIGCOMM 2013 Conference on SIGCOMM, SIGCOMM ’13, page 123–134, New York, NY, USA. Association for Computing Machinery.