基于SM-PEPA可生存系统认知模型及量化分析

发布时间：2018-05-21 12:47

本文选题：认知模型 + 随机进程代数　；参考：《华中科技大学学报(自然科学版)》2015年05期

【摘要】：为解决可生存系统的自主管理和维护问题以保持关键服务的持续提供,提出一种基于SM-PEPA的系统认知形式化模型及量化分析方法.首先建立了具有分层认知能力的可生存系统认知模型,即服务认知子层、网络认知子层和接入认知子层,然后提出了一种由策略库发起驱动基于认知环MDE的认知单元自管理模型,并通过可生存系统的生存状态转换图,构建了可生存系统认知性能分析框架.在此基础上,结合对可生存系统认知能力的形式化描述,将其转化为一个半马尔可夫过程,对模型进行了量化分析.仿真试验分析了不同参数变化对系统可生存性的影响,试验结果验证了所提方法的有效性和合理性.
[Abstract]:In order to solve the problem of autonomous management and maintenance of survivable systems and to maintain the continuous provision of critical services, a formal model of system cognition based on SM-PEPA and a quantitative analysis method are proposed. Firstly, a hierarchical cognitive model of survivable system is established, that is, service cognitive sub-layer, network cognitive sub-layer and access cognitive sub-layer. Then, a cognitive unit self-management model based on cognitive loop MDE is proposed, which is initiated by strategy library, and a framework for cognitive performance analysis of survivable systems is constructed by means of the survival state transition diagram of survivable systems. On this basis, combining with the formal description of the cognitive ability of survivable systems, the model is transformed into a semi-Markov process, and the model is analyzed quantitatively. The effects of different parameters on the survivability of the system are analyzed by simulation experiments. The experimental results verify the validity and rationality of the proposed method.
【作者单位】：哈尔滨理工大学计算机科学与技术学院;哈尔滨师范大学计算机科学与信息工程学院;
【基金】：国家自然科学基金资助项目(61403109,61202458) 高等学校博士学科点专项科研基金资助项目(20112303120007)
【分类号】：TP393.08

【参考文献】