[1]吴清亮,陶军,李鲸,等.基于神经网络逆系统的鲁棒主动队列管理算法[J].东南大学学报(自然科学版),2005,35(6):848-852.[doi:10.3969/j.issn.1001-0505.2005.06.004]
 Wu Qingliang,Tao Jun,Li Jing,et al.Robust algorithm for active queue management based on ANN inverse system[J].Journal of Southeast University (Natural Science Edition),2005,35(6):848-852.[doi:10.3969/j.issn.1001-0505.2005.06.004]
点击复制

基于神经网络逆系统的鲁棒主动队列管理算法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
35
期数:
2005年第6期
页码:
848-852
栏目:
计算机科学与工程
出版日期:
2005-11-20

文章信息/Info

Title:
Robust algorithm for active queue management based on ANN inverse system
作者:
吴清亮 陶军 李鲸 姚婕
东南大学计算机网络和信息集成教育部重点实验室, 南京 210096; 东南大学计算机科学与工程系, 南京 210096
Author(s):
Wu Qingliang Tao Jun Li Jing Yao Jie
Key Laboratory of Computer Networks and Information Integration of Ministry of Education, Southeast University, Nanjing 210096,China
Department of Computer Science and Engineering, Southeast University, Nanjing 210096, China
关键词:
主动队列管理 神经网络逆系统 鲁棒 非线性系统
Keywords:
active queue management artificial neural network inverse system Robust nonlinear system
分类号:
TP393
DOI:
10.3969/j.issn.1001-0505.2005.06.004
摘要:
通过在中间节点上使用主动队列管理策略来进行有效地拥塞控制,在保证较高吞吐量的基础上稳定地控制队列长度,从而实现了端到端的时延控制和保证QoS需求.在研究中,TCP的流量控制过程被视为二阶非线性时变系统,并通过可逆分析,证明该系统可逆,采用神经网络逆系统这种近年来发展起来的非线性鲁棒控制理论作为控制器的设计方法,设计出一种新的主动队列管理算法.仿真试验表明,这种算法的稳态和瞬态性能都优于与其具有相同实现复杂度的RED和PI算法,并且在负载扰动和参数变化时具有很强的鲁棒性.神经网络逆系统方法应用于非线性的流量控制过程中有助于系统稳定性和鲁棒性.
Abstract:
On the intermediate nodes active queue management is used for an effective congestion control policy. Base on guarantee of high throughput, it controls the queue length stabilization, then realizes the end-to-end delay control andensures QoS(quality of service)demands. In this paper, the TCP(transmission control protocol)flow process is modeled as two order nonlinear varying-time system. By analyzing the invertibility of the system, a new AQM(active queue management)algorithm based on artificial neural network inverse(ANNI)system theory that is a newly developed nonlinear theory with good robustness is proposed. The simulation results show that its stability and transient performance are superior to RED(random early detection)as well as PI algorithms. Moreover, this new AQM algorithm possesses high robustness even when network load fluctuates or system parameter changes. Artificial neural network inverse method is helpful to system stability and robustness when it is applied to nonlinear TCP flow process.

参考文献/References:

[1] Caserri C,Meo M.A new approach to model the stationary behavior of TCP connections [A].In: Proceedings of IEEE INFOCOM’00[C].Tel Aviv,Israel,CA,2000.367-375.
[2] Jacobson V.Congestion avoidance and control [J]. IEEE/ACM Transaction Networking, 1988,6(3):314-329.
[3] Floyd S,Jacobson V.Random early detection gateways for congestion avoidance [J]. IEE/ACM Transactions on Networking, 1993,1(4):397-4133.
[4] Feng W C,Kandlur D D,Saha D,et al.CSE-TR-387-99 Blue:a new class of active queue management algorithm [R].Michigan:University of Michigan Technical Reports,1999.
[5] Athuraliya S,Li V H,Low S H,et al.REM:active queue management [J]. IEEE Network,2001,15(3):48-53.
[6] Wydrowski B,Zukerman M.GREEN:an active queue management algorithm [A].In: Proceedings of ICC’02[C].New York,2002.2368-2372.
[7] Gibbens R J,Kelly F P.Distributed connection acceptance control for a connectionless network [A].In:Proc of the 16th International Teletraffic Congress [C].Edinburgh,Scotland,1999.941-952.
[8] Kunniyur S S,Srikant R.An adaptive virtual queue(AVQ)algorithm for active queue management [J]. Proc of IEEE/ACM Transaction Networking,2004,12(2):286-299.
[9] Misra V,Gong W B,Towsley D.Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED [A].In: Proceedings of the ACM/SIGCOMM’00[C].Stockholm,2000.151-160.
[10] Hollot C V,Misra V,Towsley D,et al.A control theoretic analysis of RED [A].In:Proceedings of the INFOCOMM’00[C].Tel Aviv,2000.1510-1519.
[11] Hollot C V,Misra V,Towsley D,et al.On designing improved controllers for AQM routers supporting TCP flows [A].In:Proc of IEEE INFOCOM’01 [C].Anchorage,Alaska,2001.1726-1734.
[12] Ren F,Lin C.Speed up the responsiveness of active queue management system [J]. IEICE Trans Commun,2003,E86-B(2):630-636.
[13] Dai X,He D,Zhong X,et al.MIMO system invertibility and decoupling control strategies based on ANN αth-order inversion [J]. IEE Proc Control Theory and Application,2001,148(2):125-136.
[14] 陆启韶.现代数学基础 [M].北京:北京航空航天大学出版社,1997.126-128.

相似文献/References:

[1]屠昊,纪其进,董永强.自适应虚拟队列算法AVQ的公平性研究[J].东南大学学报(自然科学版),2005,35(4):528.[doi:10.3969/j.issn.1001-0505.2005.04.007]
 Tu Hao,Ji Qijin,Dong Yongqiang.Study on fairness of adaptive virtual queue algorithm[J].Journal of Southeast University (Natural Science Edition),2005,35(6):528.[doi:10.3969/j.issn.1001-0505.2005.04.007]

备注/Memo

备注/Memo:
基金项目: 国家重点基础研究发展计划(973 计划)资助项目(2003CB314801)、高等学校博士学科点专项基金资助项目(20040286001).
作者简介: 吴清亮(1977—),男,博士生, qwu@seu.edu.cn.
更新日期/Last Update: 2005-11-20