项目名称: 海事监控传感网络中节点协同机制下的事件检测方法研究
项目编号: No.51279151
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 水利工程
项目作者: 刘克中
作者单位: 武汉理工大学
项目金额: 82万元
中文摘要: 针对海事安全保障领域传统信息采集手段存在信源种类有限、信息表达精细化程度不足、网络代价较高等局限性,在深入分析海事安全保障领域信息服务需求的基础上,系统研究海事监控传感网络中基于节点协同策略的事件检测理论与方法。研究以信息感知的精准性、全面性和及时性为目标,重点从事件信息质量可用性和节点协同有效性两个角度探讨其中关键科学问题。研究主要包括:将在分析节点级海事传感信息时空特征基础上构建事件模型,利用不完全信息条件下Markov决策理论研究事件信息质量保证下的节点协同调度策略,采用权值理论和贝叶斯方法探讨基于节点信息贡献度模型的事件检测容错方法,最后实现富含语义的海事事件信息的有效提取。研究将综合运用离散事件仿真软件和原型实验系统等手段对提出的理论、方法和算法进行验证性研究。本研究对系统形成海事监控传感网络事件检测理论,推动物联网技术在海事安全保障领域工程化应用等方面具有重要的理论和现实意义。
中文关键词: 海事监控传感网络;事件模型;节点协同;事件检测;
英文摘要: In the field of maritime safety and security, traditional information collection methods have revealed the limitations of limited types of information sources, insufficient refinement degree of information expression and high cost of networks. Thus, our focus is on the systematic study about the theory and methods of event detection based on node collaborative scheme in maritime surveillance sensor networks, which is rely on the deep analysis of the information services demands in maritime safety and security domain. Aimed at guaranteeing the accuracy, comprehensiveness and timeliness of sensory information, scientific issues focusing on two points, i.e. the availability of event information quality and effectiveness of node collaborative scheme, are majorly discussed. Furthermore, in order to fulfill the efficiency extraction of semantic maritime event information, the research includes three aspects: event modeling based on the spatio-temporal characteristics of node-leveled maritime sensory information, node co-scheduling strategy based on assuring event information quality of applying incomplete information Markov decision theory, fault-tolerant event detection methods based on node information contribution model of employing the Weighted method and Bayesian approach. Finally, our work synthetically utilizes
英文关键词: Maritime surveillance sensor networks;Event model;Node collaborative scheduling;Event detetion;