通信学报 ›› 2016, Vol. 37 ›› Issue (9): 111-120.doi: 10.11959/j.issn.1000-436x.2016183

• 学术论文 • 上一篇    下一篇

主动式复杂事件处理方法的研究

耿少峰1,2,王永恒1,李仁发1,张佳2   

  1. 1 湖南大学信息科学与工程学院,湖南 长沙 410082
    2 集美大学计算机工程学院,福建 厦门 361021
  • 出版日期:2016-09-25 发布日期:2016-09-28
  • 基金资助:
    国家自然科学基金资助项目;国家自然科学基金资助项目;福建省教育厅基金资助项目

Research of proactive complex event processing method

Shao-feng GENG1,2,Yong-heng WANG1,Ren-fa LI1,Jia ZHANG2   

  1. 1 College of Information Science and Engineering,Hunan University,Changsha 410082,China
    2 College of Computer Engineering,Jimei University,Xiamen 361021,China
  • Online:2016-09-25 Published:2016-09-28
  • Supported by:
    The National Natural Science Foundation of China;The National Natural Science Foundation of China;Foundation of Fujian Educational Committee

摘要:

在CPS的应用背景下,对传感器和控制设备产生的不确定性事件流进行分析和处理得出高层事件,然后在此基础上引入适应性动态贝叶斯网络和并行马尔可夫决策过程模型来支持主动式的复杂事件处理。针对大型CPS中马尔可夫决策过程存在的状态数量巨大的问题,引入状态划分和报酬分解的方法来进行并行优化。在模拟交通网络的环境中,实验结果显示所提方法能有效地处理事件流,并具有良好的可扩展性。

关键词: 信息物理融合系统, 大数据, 复杂事件处理, 马尔可夫决策过程

Abstract:

Based on the preliminary analysis results of the indeterminate event stream that generated by the sensors and control purpose equipment of CPS,the adaptive dynamic Bayesian network and parallel Markov decision process model were used to support the proactive complex event processing.In order to resolve the vast state space issue of Markov decision process for large CPS,states partition and reward decomposition methods were proposed to parallel the decision making process.The experimental result based on the simulation of traffic network shows that proposed method can process event stream effectively and has favorable scalability.

Key words: CPS, big data, complex event processing, Markov decision process

No Suggested Reading articles found!