Author:
Li Yanxin,Liu Shangkun,Li Jia,Zheng Weimin
Abstract
<abstract><p>This paper discusses the congestion control challenges in a network employing multi-bottleneck Transmission Control Protocol/Active Queue Management (TCP/AQM). The study specifically focuses on networks characterized by input nonlinearity and unknown disturbances. We regard the network as a whole, and consider the influence between multiple nodes and unknown disturbance, a dynamic model of multi-bottleneck network is established. And the impact of dead zone and saturation on the system is taken into account for the first time in the model, the builded TCP/AQM model is more practicable. Based on the characteristics of fuzzy logic systems (FLS), combined with backstepping technology and Lyapunov function, an adaptive congestion control algorithm is designed to make full use of the link resources of each node and improve the network utilization. Ultimately, the proposed algorithm's efficacy and superiority are substantiated through simulation.</p></abstract>
Publisher
American Institute of Mathematical Sciences (AIMS)
Reference39 articles.
1. V. Jacobson, Congestion avoidance and control, ACM SIGCOMM Comput. Commun. Rev., 18 (1988), 314–329. https://doi.org/10.1145/52325.52356
2. M. Mathis, J. Mahdavi, S. Floyd, A. Romanow, TCP selective acknowledgment options, Tech. Rep., 1996.
3. S. Ryu, C. Rump, C. Qiao, Advances in internet congestion control, IEEE Commun. Surv. Tut., 5 (2003), 28–39. https://doi.org/10.1109/COMST.2003.5342228
4. G. Chatranon, M. A. Labrador, S. Banerjee, A survey of tcp-friendly router-based AQM schemes, Comput. Commun., 27 (2004), 1424–1440. https://doi.org/10.1016/j.comcom.2004.05.001
5. T. A. N. Nguyen, S. Gangadhar, J. P. Sterbenz, Performance evaluation of TCP congestion control algorithms in data center networks, In: Proceedings of the 11th International Conference on Future Internet Technologies, 2016, 21–28. https://doi.org/10.1145/2935663.2935669