一种基于异常点检测的电信网络性能监控策略

于艳华; 宋俊德

doi:10.3724/SP.J.1146.2008.01196

一种基于异常点检测的电信网络性能监控策略

doi: 10.3724/SP.J.1146.2008.01196

基金项目:

十一五国家科技支撑计划(2006BAH02A03)资助课题

计量
- 文章访问数: 3133
- HTML全文浏览量: 71
- PDF下载量: 1024
- 被引次数: 0
出版历程
- 收稿日期: 2008-09-22
- 修回日期: 2009-05-14
- 刊出日期: 2009-09-19

A Mechanism of Telecommunication Network Performance Monitoring Based on Anomaly Detection

摘要

摘要: 电信网络运营维护方式在由面向网络向面向客户转变，电信网络性能管理也需要由被动地监测向主动地监控转变。主动性能监控机制通过检测代表性能降质的异常点来实现故障的快速恢复。该文提出一种基于支持向量机时间序列预测(Support Vector Machines, SVM)和相关置信区间的异常点检测机制。另外，提出一种新的支持向量机时间序列建模的自由参数的选取方法，并给出计算机实现过程。理论分析和实验结果表明了该自由参数选取方法的正确性和异常点检测机制的有效性。
- 电信网络; 主动性能监控; 时间序列预测; 支持向量机; 白噪声; 置信区间
Abstract: With operation and maintenance mode of telecommunication network changing from network oriented to subscriber oriented, network performance management should also be changed from passive monitoring to proactive monitoring. Proactive Performance Monitoring (PPM) enables a fast fault correction by detecting anomalies designating performance degradation. In this paper, a novel anomaly detection approach is proposed taking advantage of time series prediction and associated confidence interval based on Support Vector Machines (SVM). In addition a novel meta-parameters selection approach is proposed by checking if the training residual is white noise. Theoretical analysis and experimental results verify the correctness of the meta-parameter selection approach and the effectiveness of anomaly detection mechanism.

HTML全文

参考文献(1)

Feather F and Maxion R. Fault detection in an ethernetnetwork using anomaly signature matching. Proc. ACMSIGCOMM, San Francisco, CA, 1993: 279-288.[2]Ho L L, Cavuto D J, and Papavassiliou S, et al.. Adaptiveand automated detection of service anomalies in transactionorientedWAN's: network analysis, algorithms,implementation, and deployment [J].IEEE Journal ofSeletected Areas in Communications.2000, 18(5):744-757[3]Box G E P, Jenkins G M, and Reinsel G C. Time SeriesAnalysis: Forecasting and Control [M]. 3rd ed. EnglewoodCliffs: Prentice Hall, 1994, chapter 3-chapter 4.[4]Methaprayoon K, Lee Wei-jen, and Rasmiddatta S, et al..Multistage Artificial neural network short-Term loadforecasting engine with front-end weather forecast [J].IEEETransactions on Industry Applications.2007, 43(6):1410-1416[5]Muller K R and Smola A. Prediction time series with supportvector machines. Proc. ICANN, Lausanne, Switzerland, 1997:999-1004.[6]Kaheil Y H, Rosero E, and Gill M K, et al.. Downscaling andforecasting of evapotranspiration using a synthetic model ofwavelets and support vector machines[J].IEEE Transactionson Geoscience and Remote Sensing.2008, 46(9):2692-2707[7]Shi Zhi-wei and Han Min. Support vector echo-state machinefor chaotic time-series prediction [J].IEEE Transactions onNeural Networks.2007, 18(2):359-372[8]Vapnik V. The Nature of Statistical Learning Theory [M].New York: Springer-Verlag, 1995, Chapter 6.[9]邓乃扬, 田英杰. 数据挖掘中的新方法支持向量机[M].北京: 科学出版社, 2004: 143-161.Deng Nai-yang and Tian Ying-jie. The New Method of DataMining: Support Vector Machines [M]. Beijing: Science Press,2004: 335-341.[10]朱树先, 张仁杰. 支持向量机核函数选择对面部特征识别的作用[J]. 光学技术, 2008, 34(6): 902-904.Zhu Shu-xian and Zhang Ren-jie. Kernel selection for supportvector machines used in face feature recognition [J]. OpticalTechnique, 2008, 34(6): 902-904.[11]Gold C and Sollich P. Model selection for support vectormachine classification [J].Neurocomputing.2003, 55(1-2):221-249[12]Smola A.[J].Murata N, and Sch.lkopf B, et al.. Asymptoticallyoptimal choice of -loss for support vector machines. Proc.ICANN, Skovde, Sweden.1998,:-[13]Cherkassky V and Ma Yun-qian. Practical selection of SVMparameters and noise estimation for SVM regression [J].Neural Networks.2004, 17(1):113-126[14]Sch.lkopf B, Bartlett P, and Smola A, et al.. Support vectorregression with automatic accuracy control. Proceedings ofICANN'98, Berlin, 1998: 111-116.

施引文献

资源附件(0)

访问统计