Fuzzy Three Time Scale Congestion Controller
Although loss rate, length of the queues in the routers and throughput are affected by self similar property of traffic, but classic congestion control algorithms work in short time scales and do not consider self similarity and long range (time) dependency phenomenon of data. To profit these properties, researchers have proposed several methods. Multi time scale congestion control is one of the successful ways to adapt with self similar traffic and predict network status. In this research, a three part structure has been implemented in which second and third parts take advantage of fuzzy engines. Results show throughput improvement in case of using fuzzy three time scale controller instead of a two time scale controller or a classic controller such as New Reno.
KeywordsSelf similarity multi time scale long range dependency fuzzy congestion control
Unable to display preview. Download preview PDF.
- 2.Crovella, M., Bestavors, A.: Self similarity in world wide web traffic: Evidence and possible causes. In: International Conference on Measurement and Modeling of Computer Systems (ACM SIGMETRICS). ACM, Philadelphia (1996)Google Scholar
- 6.Lu, J., Ruan, Q., Ni, R.: Fractal-based multiple time scale TCP-friendly congestion control for multimedia Streaming. In: 18th Canadian Conference on Electrical and Computer Engineering, Saskatchewan (2005)Google Scholar
- 7.Lu, J., Ni, R.: Media Streaming TCP-Friendly Congestion Control Using Multiple Time Scale Prediction. In: Second International Conference on Innovative Computing, Information and Control, Kumamoto, vol. 1(1), pp. 535–540 (2007)Google Scholar
- 8.Hagivara, T., Majima, H., Matsuda, T., Yamamoto, M.: Impact of Round Trip self-similarity on TCP performance. In: 10th International Conference on Computer Communications and Networks (2001)Google Scholar
- 9.Mohtashamzadeh, M., Soryani, M., Fathy, M.: Fuzzy two time-scale congestion control algorithm. In: International Conference on Computational Intelligence, Communication Systems and Networks. IEEE, Indore (2009)Google Scholar
- 10.Omnetpp V3.3 simulator, http://www.omnetpp.com
- 11.Tian, X., Wu, H., Ji, C.: A unified framework for understanding network traffic using independent wavelet models. In: Proceedings of IEEE Infocom 2002, New York (2002)Google Scholar