压缩感知的多参数链路故障定位算法  

Multi-parameters Link Failure Localization Algorithm Based on Compressive Sensing

作　　者：王汝言[1] 吴晴[1] 熊余[1,2] 谢雨[1] 赵莹[1] 

机构地区：[1]重庆邮电大学重庆市光纤通信技术重点实验室,重庆400065 [2]重庆大学计算机学院,重庆400030

出　　处：《电子与信息学报》2013年第11期2596-2601,共6页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(60972069;61001105);重庆市自然科学基金重点项目(2011BA2041);重庆市教委科学技术研究项目(KJ110531);重庆市高校优秀人才支持计划(2011-29)资助课题

摘　　要：为了提高故障定位性能,降低单一判别参数在单位过程中的约束,该文提出一种基于压缩感知和信息熵差的多参数链路故障定位算法。该算法首先利用贝叶斯网络进行快速故障预测,其次引入参数故障覆盖范围,利用压缩感知进行故障筛选,最后定义参数故障信息熵差完成根源故障定位。仿真结果表明,该算法预测出的故障集合具有可压缩性,筛选后的故障集合保留了真实故障,定位时具有较高的故障检测率和较低的故障误检率。To improve the performance and decrease the constraints of fault localization with single distinguish parameter, a multi-parameters link failure localization algorithm is proposed based on compressive sensing and entropy difference. Firstly it makes a fast fault prediction by Bayesian network, then it introduces a parameter named fault coverage and selects probable link failure using compressive sensing, finally defines fault information entropy difference and obtains the root fault based on the parameter. The simulation results show that the predicted fault set can be compressed and the selected probable fault set contains the true fault, meanwhile the proposed algorithm achieves high detection rate and low false positive rate.

关 键 词：压缩感知 故障定位 贝叶斯网络 信息熵 

分 类 号：TN915[电子电信—通信与信息系统]