基于T-S模糊观测器的网络拥塞控制算法  

Network Congestion Control Algorithm Based on T-S Fuzzy Observer

作　　者：王宏伟[1] 于驰[1] 井元伟[2] 

机构地区：[1]东北大学秦皇岛分校,河北秦皇岛066004 [2]东北大学信息科学与工程学院,辽宁沈阳110004

出　　处：《东北大学学报（自然科学版）》2010年第4期461-464,468,共5页Journal of Northeastern University(Natural Science)

基　　金：国家高技术研究发展计划项目(2004AA412030)

摘　　要：针对TCP(传输控制协议)网络的拥塞控制问题,设计了基于模糊观测器的主动队列管理算法.对非线性TCP/IP网络拥塞控制系统建立了T-S模糊模型,通过选取适当的模糊规则和隶属函数来提高拥塞控制系统的性能,并给出了理论性证明.仿真结果表明所设计的控制器对活动的TCP连接数、链路带宽及往返时延的不确定性具有很强的稳定性和鲁棒性.To solve the problem of network congestion control in accordance to TCP（transmission control protocol）,an active queue management（AQM） algorithm was designed on the basis of fuzzy observer.With a T-S fuzzy model developed for the nonlinear TCP/IP network congestion control system,the performance of the network system was improved by choosing appropriate fuzzy rules and membership functions,and the stability of the system was proved theoretically.Simulation results in different scenarios demonstrated that the proposed controllers are highly stable and robust in respect to the uncertainties of the number of active TCP sessions,link capacity and the round-trip time（RTT）.

关 键 词：TCP/IP网络 拥塞控制 T-S模糊模型 主动队列管理(AQM) 鲁棒性 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]