大尺度IP骨干网流量矩阵估计方法研究.pdf

4 Vol .39 No. 4 2011 4 ACTA ELECTRONICA SINICA Apr. 2011 IP 1, 2 1 1 1 1 蒋定德 , 王兴伟 , 郭 磊 , 许争争, 陈振华 ( 1. , 110004; 2. , 610054) : , . , , . IP ; RBF( Radial Basis Function) , TMRI (Traffic Matri Recurrence Inference) .TMRI RBF , RBF , . , , , TMRI . TMRI , , . : ; ; RBF ; ; : TP393 : A : ( 2011) Approach of Traffic Matrix Estimation in Largescale IP Backbone Networks 1, 2 1 1 1 1 JIANG Dingde , WANG Xingwei , GUO Lei , Xu Zhengzheng , CHEN Zhenhua ( 1. College of Inf ormation Science and Engineering,Northeastern University, Shenyang, Liaoning 110004, China; 2. KeyLaboratory of Broadband ptical Fiber Transmission and Communication Networks, UESTC, Chengdu, Sichuan 610054, China) Abstract: Traffic matri estimation is an interesting research problem at present. Network operators use it to conduct load balancing , route optimization, traffic detecting, network dimensioning and so on. However , the highly illpose nature of traffic matri estimation itself makes it being a challenging research subject to estimate accurately traffic matri . This paper studies traffic matri estimation in largescale IP backbone networks. Based on RBF ( radial basis function) neural network, a novel estimation method, namely TMRI ( traffic matri recurrence inference) , is proposed. TMRI e ploits the powerful modeling ability of RBF neural net work to model traffic matri estimation. The illpose nature of this problem will be overcome in the process of training the RBF neu ral network. Accordingly, the comple process of mathemat