移动通信网络的中性集故障诊断方法研究  

Research on neutral set fault diagnosis method for mobile communication networks

作　　者：张婷 齐小刚 ZHANG Ting;QI Xiaogang(School of Mathematics and Statistics,Xidian University,Xi’an 710071,China;Xidian-Ningbo Information Technology Institute,Ningbo 315200,China)

机构地区：[1]西安电子科技大学数学与统计学院,陕西西安710071 [2]西安电子科技大学宁波信息技术研究院,浙江宁波315200

出　　处：《智能系统学报》2020年第5期864-869,共6页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金项目(61572435,61877067);教育部−中国移动联合基金项目(MCM20170103);西安市科技创新项目(201805029YD7CG13-6);宁波市自然科学基金项目(2016A610035;2017A610119)。

摘　　要：随着故障诊断技术从面向网络设备逐渐向面向用户、面向业务的转变,故障识别能力和故障处理能力的不断提高,快速、准确地检测重大故障并及时收集重大故障信息,对缩短故障时长、提高工作效率、提升网络服务水平都具有重大意义。基于中性集故障诊断方法,本文提出一种故障特征权重和阶段数据权重的计算方法,并将此方法应用于移动通信网络的话音业务的故障诊断过程中。根据收集到的数据的统计分析结果,判断移动通信网络中未知故障样本的故障类型。通过举例分析,验证了本文所提出的故障特征权重和阶段数据权重设计方法的优越性。With the change in network fault diagnosis technology—from equipment-oriented to user-oriented and business-oriented,with continuous improvement of fault identification ability and fault handling ability—rapid and accurate detection of major faults and timely collection of major fault information are of great significance to shortening the fault time,improving work efficiency,and raising the network service level.Based on the neutral set fault diagnosis method,this paper proposes a method for calculating the fault feature weight and phase data weight.Then,the method is applied for the fault diagnosis of voice service in mobile communication networks.The fault magnitude of unknown fault samples in mobile communication networks is judged according to the statistical analysis result of the collected data.The superiority of the proposed design method for fault feature weight and stage data weight is verified through the analysis of an example.

关 键 词：故障诊断 中性集 不确定性 移动通信网络 话音业务 特征权重 阶段数据权重 故障类型 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]