基于异构传感网的事件跟踪和查询技术

The heterogeneous sensor network, composed of sensors having different functions, plays an important role in the sensing system of Internet of Things. This project studies the problem of event tracking and query in the heterogeneous sensing environment, including real-time query、 continuous target location tracking、multi-attribute query. At present, the traditional tracking and querying techniques mainly study the single target tracking and single attribute query, which have been unable to adapt to the tracking and query requirements involved in the study of this project. This project considers the factors of moving objects similarity intensity and regional popularity, and intends to propose a continuous target location prediction mechanism to ensure the real-time query of single attribute in the heterogeneous sensing environment, which is expected to be directly extended to multiple attributes; studies the common distribution trend of the sensors having different functions, and combining with the cluster technology, intends to construct an energy-efficient spatial index mechanism that is appropriate for multi-attribute query in the heterogeneous sensing environment; studies the skewed distribution state of mobile objects, and with the minimum energy principle as the cluster similarity constraint, intends to propose a shortest routing and threshold filtering mechanism that is suitable for the effective filter and comparison of multi-attribute aggregated value in the heterogeneous sensing environment; intends to construct a real-time information tracking and query system to verify the correctness and feasibility of the proposed theory. This project aims to solve the problem of event tracking and query in the heterogeneous sensing environment, which has the significance of promoting the solving of the other tracking and query problems in the heterogeneous sensing environment.

由不同功能传感器组成的异构传感网在物联网感知系统中起着重要作用。本项目研究异构感知环境中事件的跟踪和查询问题，包括实时查询、连续目标位置跟踪、多属性查询。目前传统的跟踪和查询技术主要研究单目标的跟踪和单属性的查询，已无法适应本项目研究中涉及的跟踪和查询需求。本项目考虑移动对象相似性强度和区域流行性因素，拟提出异构感知环境中保证单属性实时查询的连续目标位置预测机制，该机制有望直接扩展至多属性；研究不同功能传感器的共同分布趋势，结合聚类技术，拟构建异构感知环境中适宜多属性查询的能量有效空间索引机制；分析移动对象的偏斜分布状态，以能量最小原则作为聚类相似性约束，拟提出异构感知环境中适宜多属性集成值有效过滤与比较的最短路由机制；拟建立实时信息跟踪和查询系统验证所提出理论的正确性和可行性。本项目致力于解决异构感知环境中事件的跟踪和查询问题，这对促进异构感知环境中其它跟踪和查询问题的解决有重要意义。

传统的跟踪和查询技术主要研究单目标的跟踪和单属性的查询，已无法适应本项目研究中涉及的跟踪和查询需求。项目主要对“基于异构感知数据的事件跟踪和查询问题”进行深入的研究，其主要内容包括：（1）雾支持的工业无线传感网多属性top-k查询，（2）云边融合环境下基于合作性云边缓冲的物联网搜索，（3）云边融合环境中连续对象区域的探测。项目降低了通信代价，显著提升了跟踪查询时效、跟踪查询效率和能量有效性。.项目完成论文8篇，其中CCF B类会议论文1篇，SCI一区论文2篇，二区论文1篇，三区论文2篇，EI检索会议论文2篇。论文数量达到申请书中的目标。.该项目解决了物联网中对象搜索、连续对象跟踪探测等核心问题，使得物联网对象搜索跟踪有了更好的性能和更广阔的应用范围。基于现有成果开发的物联网跟踪搜索系统可以被用在车联网、云边融合环境中异常探测等重要研究领域。