基于模拟退火和广义反演的流量矩阵估计  

Simulated Annealing and Generalized Inference-based traffic matrix estimation

作　　者：廖建平[1] 

机构地区：[1]衢州职业技术学院信息工程学院,浙江衢州324000

出　　处：《微计算机信息》2012年第10期463-465,共3页Control & Automation

摘　　要：研究时变网络条件下大尺度IP骨干网络流量矩阵估计问题,通过使用修改的模拟退火方法来捕捉流量矩阵特征,将流量矩阵估计描述成欧氏距离和马氏距离下的最优化过程,能成功克服流量矩阵估计的病态特性,获得精确的估计值,仿真结果表明,该估计算法估计精度和显著的性能改善。This paper studied traffic matrix estimation in the large-scale IP networks. By using modified simulated annealing method to capture the characteristics of traffic matrix and describing the traffic matrix estimation into an optimal process under the Euclid distance and Mahalanobis distance, the ill-posed nature of this problem could successfully be overcome and the accurate estimation could be obtained. Simulation results show that estimation algorithm proposed holds the higher estimation accuracy and evident perfor- mance improvement.

关 键 词：关键字 流量矩阵估计 网络流量 模拟退火 广义反演 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]