光子学报, 2014, 43 (7): 0706023, 网络出版: 2014-08-18  

端到端网络流量的混合估计方法

Mixed Estimation Approach to EndtoEnd Network Traffic
作者单位
东北大学 信息科学与工程学院, 沈阳 110819
摘要
利用主成分分析法获取端到端网络流量的主要特征分量并获得其初始估计结果.为克服其初值敏感性将估计结果作为遗传算法的初始值、链路流量估计偏差作为遗传算法的适应度函数, 通过构建合适的交叉和变异概率函数来控制遗传算法的交叉和变异过程.采用合适的约束迭代函数,利用遗传算法通过迭代寻优获得端到端流量的估计结果, 仿真结果表明所提出的方法是可行的.
Abstract
Principal component analysis was exploited to extract the principal features of endtoend network traffic and to attain the initial estiamtion results. This results are taken as the prior value of genetic algorithm to overcome its sensitiveness to the prior value. The estiamtion biases of link traffic is regarded as the fitness function of genetic algorithm. The crossover and mutation probability functions are built to control its corossover and mutation processes. The appropriate iterative funtion with contraints is built. The genetic algorithm is used to attain the endtoend traffic estimation results in the iterative way. Simulation results show that the proposed method is feasible.
参考文献

[1] WANG X, ZHANG D. A novel method to estimate IP traffic matrix[C]. Proceedings of WiCOM′10, 2010: 1-4.

[2] CAI Ting, HUANG Shanguo, LI Xin, et al. Dynamic survivable mapping algorithm based on ant colony optimization in IP over WDM networks[J]. Acta Photonica Sinica, 2012, 41(12): 1400-1404.

[3] JIANG Dingde, XU Zhengzheng, NIE Laisen, et al. An approximate approach to endtoend traffic in communication networks[J]. Chinese Journal of Electronics, 2012, 21(4): 705-710.

[4] GUAN Aihong, WANG Boyun, FU Hongliang, et al. A deflection routing mechanism based on priority and burst segmentation in optical burst switching networks[J]. Acta Photonica Sinica, 2012, 41(2): 127-132.

[5] AKGUL T, BAYKUT S, KANTARCI M, et al. Periodicitybased anomalies in selfsimilar network traffic flow measurements[C]. Proceedings of TIM′11, 2011, 60(4): 1358-1366.

[6] JIANG Dingde, QIN Wenda, TANG Qingyi, et al. An estimation approach to traffic matrix in optical networks based on network tomography[J/OL]. [20131104]. http://www.photon.ac.cn/CN/abstract/abstract20262.shtml.

[7] TEBALDI C, WEST M. Bayesian inference on network traffic using link count data[J]. Journal of American Statistics Association, 1998, 93(442): 557-576.

[8] JUVA I, KUUSELA P, VIRTAMO J. A case study on traffic matrix estimation under Gaussian distribution[C]. In Proceedings of NTS′04, 2004: 49-60.

[9] VATON S, BEDO J. Network traffic matrix: How can one learn the prior distributions from the link counts only[C]. In Proceedings of ICC′04, 2004: 2138-2142.

[10] SOULE A, LAKHINA A, TAFT N, et al. Traffic matrices: balancing measurements, inference and modeling[C]. In Proceedings of SIGMETRICS′05, 2005, 33(1): 362-373.

[11] ZHANG Y, ROUGHAN M, DUFFIELD N, et al. Fast accurate computation of largescale IP traffic matrices from link loads[C]. In Proceedings of SIGMETRICS′03, 2003, 31(3): 206-217.

[12] TAKEDA T, SHIONOTO K. Traffic matrix estimation in largescale IP networks[C]. In Proceedings of LANMAN′10, 2010: 1-6.

[13] JIANG Dingde, XU Zhengzheng, CHEN Zhenhua, et al. Joint timefrequency sparse estimation of largescale network traffic[J]. Computer Networks, 2011, 55(10): 3533-3547.

[14] JIANG Dingde, XU Zhengzheng, XU Hongwei, et al. An approximation method of origindestination flow traffic from link load counts[J]. Computers and Electrical Engineering, 2011, 37(6): 1106-1121.

蒋定德, 赵祖耀, 许宏伟, 王兴伟. 端到端网络流量的混合估计方法[J]. 光子学报, 2014, 43(7): 0706023. JIANG Dingde, ZHAO Zuyao, XU Hongwei, WANG Xingwei. Mixed Estimation Approach to EndtoEnd Network Traffic[J]. ACTA PHOTONICA SINICA, 2014, 43(7): 0706023.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!