期刊名称：International Journal of Security and Its Applications
印刷版ISSN：1738-9976
出版年度：2015
卷号：9
期号：10
页码：1-20
DOI：10.14257/ijsia.2015.9.10.01
出版社：SERSC
摘要：Integrated Modular Avionics (IMA) uses task synthesis, function fusion and resource integration to achieve the goal of low cost, high efficiency, high efficacy, high performance and high reliability. However, safety issue is caused in the system integration process. In this paper, firstly, we use data mining technology to describe resource-layer safety model, function-layer safety model and task-layer safety model; secondly, we proposed an efficient bicluster mining algorithm: LowCluster, to effectively mine all the maximal constant row biclusters with low usage rate in real-valued function- resource matrix for IMA safety analysis. In LowCluster algorithm, a sample weighted graph is constructed firstly, it includes all resource collections between both samples which meet the definition of low usage rate; then, all the maximal constant row biclusters with low usage rate are mined using sample-growth and depth-first method in the sample weighted graph. In order to improve the mining efficiency, LowCluster algorithm uses pruning strategy to ensure the mining of maximal bicluster without candidate maintenance. The experimental results show that LowCluster algorithm is more efficient than traditional constant row biclustering algorithm, and using our proposed LowCluster algorithm can find the error reason when executing more functions, which will help to improve system safety analysis.
