|
|
- 2016
互联网远程实时控制系统短时间窗口往返时延测量、分析与建模
|
Abstract:
为了得到准确地描述互联网远程实时控制系统时延的分布模型,对短时间窗口往返时延(round？？trip time, RTT)进行了测量、统计和分析,提出了一种短时窗RTT时延混合威布尔(Weibull)模型。该混合模型满足网络控制系统的实时性需求,具备对短期非平稳的短时间窗口RTT时延样本随机聚类特性的描述能力,并利用期望最大化(expectation？？maximization, EM)算法估计模型的混合分量密度及混合分量模型参数。实验结果表明:使用二重混合威布尔模型建模短时窗RTT时延样本能够很好地反映时延的随机聚类特性,K？？S检验显示该模型对样本匹配效果评价可接受,该模型为互联网远程控制系统的优化控制提供了一种更准确的参考模型。
A mixture Weibull model for short window RTT delay is proposed to obtain an accurate model to describe the delay distribution of real？？time Internet remote control systems by measuring and analyzing short window round？？trip delay(RTT). The model meets real？？time requirements of network control systems, has the ability to describe the randomized clustering feature for non？？stationary short window RTT delay, and uses the expectation？？maximization(EM) algorithm to estimate component densities and parameters of the model. Experimental results confirm the efficiency and precision of 2？？mixture Weibull model for the short window RTT delay, and a K？？S test shows an acceptable result that the model matches the sample. It is concluded that the model provides a more accurate reference model for optimal control of Internet real？？time remote control systems
| [1] | ［1］CIUCU F, BURCHARD A, LIEBEHERR J. A network service curve approach for the stochastic analysis of networks ［C］∥Proceedings of the 2005 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems. New York, USA: ACM, 2005: 279？？290. |
| [2] | ［2］韩悦, 刘增基, 姚明？J. 基于指数上鞅的统计端到端时延分析 ［J］. 计算机学报, 2012, 35(10): 2016？？2022. |
| [3] | HAN Yue, LIU Zengji, YAO Mingwu. Exponential supermartingale for the stochastic analysis of end？？to？？end delay ［J］. Chinese Journal of Computers, 2012, 35(10): 2016？？2022. |
| [4] | ［3］NG T E, ZHANG H. Predicting internet network distance with coordinates？？based approaches ［C］∥Proceedings of the Twenty？？First Annual Joint Conference of the IEEE Computer and Communications Societies. Piscataway, NJ, USA: IEEE, 2002: 170？？179. |
| [5] | ［4］ZHANG B, NG T S E, NANDI A, et al. Measurement based analysis, modeling, and synthesis of the internet delay space ［C］∥Proceedings of the 6th ACM SIGCOMM Conference on Internet Measurement. New York, USA: ACM, 2006: 85？？98. |
| [6] | ［6］林川, 赵海, 毕远国, 等. 互联网网络时延特征研究 ［J］. 通信学报, 2015(3): 167？？178. |
| [7] | LIN Chuan, ZHAO Hai, BI Yuanguo, et al. Research on network delay of Internet ［J］. Journal on Communications, 2015(3): 167？？178. |
| [8] | ［7］李超, 赵海, 张昕, 等. Internet中支配延迟的特征行为研究 ［J］. 电子学报, 2008, 36(6): 1063？？1067. |
| [9] | LI Chao, ZHAO Hai, ZHANG Xin, et al. Research on the characteristic behavior of internet dominant delay ［J］. Acta Electronica Sinica, 2008, 36(6): 1063？？1067. |
| [10] | ［8］MARKOPOULOU A, TOBAGI F, KARAM M. Loss and delay measurements of internet backbones ［J］. Computer Communications, 2006, 29(10): 1590？？1604. |
| [11] | ［11］ZHANG L, GOVE J H, LIU C, et al. A finite mixture of two Weibull distributions for modeling the diameter distributions of rotated？？sigmoid, uneven？？aged stands ［J］. Canadian Journal of Forest Research, 2001, 31(9): 1654？？1659. |
| [12] | ［13］HERNANDEZ J A, PHILLIPS I W. Weibull mixture model to characterise end？？to？？end Internet delay at coarse time？？scales ［J］. Proceedings of Communications, 2006, 153(2): 295？？304. |
| [13] | ［9］GEARY R. Relative efficiency of count of sign changes for assessing residual autoregression in least squares regression ［J］. Biometrika, 1970, 57(1): 123？？127. |
| [14] | ［10］REDNER R A, WALKER H F. Mixture densities, maximum likelihood and the EM algorithm ［J］. SIAM Review, 1984, 26(2): 195？？239. |
| [15] | ［12］JIANG R, MURTHY D. Mixture of Weibull distributions？？parametric characterization of failure rate function ［J］. Applied Stochastic Models and Data Analysis, 1998, 14(1): 47？？65. |
| [16] | ［5］PADMANABHAN V N, SUBRAMANIAN L. An investigation of geographic mapping techniques for internet hosts ［C］∥Proceedings of the ACM SIGCOMM Computer Communication Review. New York, USA: ACM, 2001: 173？？185. |
