一种基于AADL错误模型的软件安全性分析技术研究

Analyzing Software Safety with AADL Error Model

针对软件安全性问题,提出一种新的软件安全性分析方法。首先探索将软件组件AADL错误模型转化为马尔科夫链模型,计算组件处于不同危害级别状态概率,分析组件安全性。并且进一步,根据AADL错误模型组合实现定义,由其内部组件安全状态概率计算分析软件系统的安全性,避免状态爆炸问题。最后,以飞控系统软件为例,对算法进行实例验证。

We propose a new model-based software safety analysis method, which can analyze software safety early in the software design phase.Firstly we define the rules for transforming the AADL error model into expanded Mark-ov chain;the probabilities of different level component′s safety states can be calculated with Markov model.AADL can define software system error behavior model in terms of the subcomponent error models, then safety of software system can be got directly in terms of subcomponent safeties.Finally an example is given to explain the use of the measurement method.