文献类型：期刊文献

中文题名：Reliability Estimation for Component-Based Software Using General Masking Grouped Data

英文题名：Reliability Estimation for Component-Based Software Using General Masking Grouped Data

作者：杨剑锋 陈静 胡文生

机构：[1]Faculty of Information Engineering,Guizhou Institute of Technology;[2]School of Mathematics and Computer Science,Guizhou Normal University

第一机构：贵州理工学院

年份：2016

卷号：33

期号：6

起止页码：908-913

中文期刊名：东华大学学报：英文版

收录：Scopus

基金：Technology Foundation of Guizhou Province,China(No.QianKeHeJZi[2015]2064);Scientific Research Foundation for Advanced Talents in Guizhou Institue of Technology and Science,China(No.XJGC20150106);Joint Foundation of Guizhou Province,China(No.QianKeHeLHZi[2015]7105)

语种：中文

中文关键词：masked data;software reliability;non-homogeneous Poisson process(NHPP);maximum likelihood estimation;immune particle swarm optimization(IPSO)

外文关键词：masked data ; software reliability; non-homogeneous Poisson process （NHPP）; maximum likelihood estimation; immune particle swarm optimization （IPSO）

摘要：Masked data are the system failure data when exact component causing system failure might be unknown.In this paper,the mathematical description of general masked data was presented in software reliability engineering.Furthermore,a general maskedbased additive non-homogeneous Poisson process(NHPP) model was considered to analyze component reliability.However,the problem of masked-based additive model lies in the difficulty of estimating parameters.The maximum likelihood estimation procedure was derived to estimate parameters.Finally,a numerical example was given to illustrate the applicability of proposed model,and the immune particle swarm optimization(IPSO) algorithm was used in maximize log-likelihood function.
Masked data are the system failure data when exact component causing system failure might be unknown. In this paper, the mathematical description of general masked data was presented in software reliability engineering. Furthermore, a general masked- based additive non-homogeneous Poisson process （NHPP） model was considered to analyze component reliability. However, the problem of masked-based additive model lies in the difficulty of estimating parameters. The maximum likelihood estimation procedure was derived to estimate parameters. Finally, a numerical example was given to illustrate the applicability of proposed model, and the immune particle swarm optimization （IPSO） algorithm was used in maximize log-likelihood function.