摘要
数据异常检测与定位具有重要的实际应用价值,为了提高灵敏光纤网络异常数据检测的准确性,提了一种小波分析和改进支持向量机的高灵敏光纤网络异常数据检测方法(WA-ACO-SVM)。首先采用小波分析对高灵敏光纤网络数据进行分解,降低噪声数据被误当作异常数据的概率,然后采用蚁群优化支持向量机对高灵敏光纤网络异常数据的检测与定位进行建模,最后通过高灵敏光纤网络异常数据检测仿真试验测试其性能。结果表明,本文不仅提高了高灵敏光纤网络异常数据的检测精度,异常数据检测的时间短,具有较好的实时性。
Data anomaly detection and location has important practical application value, in order to improve the detection accuracy of the abnormal data in the optical fiber network, a new method for detecting the abnormal data of the high sensitive fiber network based on wavelet analysis and improved support vector machine is proposed. Wavelet analysis is used to decompose high sensitive fiber network data to reduce the probability which noise data is mistaken as abnormal data, and secondly, detection and localization of the abnormal data in high sensitive fiber network is modeled by using ant colony optimization support veetor machine, finally, the performance is tested by of the simulation experiment for detection and location of abnormal data in the high sensitive fiber network. The results show that the proposed method not only improves the detection accuracy of abnormal data of high sensitive fiber network, and has better realtime performance.
出处
《激光杂志》
北大核心
2016年第7期93-96,共4页
Laser Journal
基金
河南省高等学校重点科研项目计划(15A520104)
关键词
光纤网络
数据异常
检测模型
定位
optical fiber network
data anomaly
detection model
作者简介
王红霞（1978-），女（汉族），河南温县人，硕士，讲师，主要研究方向：网络安全，数据库；
刘丽（1978-），女（汉族），河南新乡人，博士，副教授，研究方向：密码学、数字水印。
