基于贝叶斯征兆解释度的链路故障定位算法  

Link failure localization algorithm based on Bayesian symptom explained degree

作　　者：王汝言[1] 吴晴[1] 熊余[1,2] 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：《计算机应用研究》2013年第3期712-714,719,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(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资助项目(2011-29)

摘　　要：针对故障和征兆关系不确定的网络中故障定位算法检测率低和误检率高的缺陷,提出了一种基于贝叶斯征兆解释度的链路故障定位算法。该算法以概率加权的二分图作为故障传播模型,通过处理贝叶斯后验概率信息,定义一种新的参数贝叶斯征兆解释度,并基于该参数对可能链路故障进行判断,得出最优故障假设集合,实现链路故障定位。理论分析和仿真实验表明,该算法具有较低的计算复杂度,且在小规模不确定网络中具有较高的故障检测率和较低的故障误检率。Aiming at the low detection rate and high false positive rate of fault localization algorithm in network of uncertainty relationship between fault and symptoms, this paper proposed a link failure localization algorithm based on Bayesian symptom explained degree. This algorithm took probabilistic weighted bipartite graph as fault propagation mode/, it defined a novel pa- rameter Bayesian symptom explained degree by handling the Bayesian posterior probability, and dealt with the possible link failure based on the parameter, then obtained the optimal fault hypothesis set and realized link failure localization. The theory analysis and simulation results show that the algorithm has lower complexity, and it has higher fault detection rate and lower false positive rate in uncertainty small size network.

关 键 词：网络生存性 故障定位 故障传播模型 贝叶斯网络 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] TP301.6[自动化与计算机技术—计算机科学与技术]