基于无线传感器网络的Skyline查询技术研究

发布时间：2018-03-28 05:15

本文选题：无线传感器网络　切入点：Skyline查询　出处：《天津理工大学》2017年硕士论文

【摘要】：无线传感器网络(WSN:Wireless Sensor Network)是分布式多跳自组织网络,并且节点的能量都是由电池提供的。现有的基于单服务器的Skyline查询算法已经不能很好应用于WSN这类分布式多跳自组织网络。针对WSN的自身特点,研究基于WSN的高效节能的Skyline查询方法有着重要的理论与应用价值。本文针对WSN中数据流的特点,分别研究了基于数据流的快照Skyline查询和连续Skyline查询处理算法,同时对数据流常用的滑动窗口技术进行了研究。本文所做的主要研究工作如下:(1)提出了WSN中的基于聚簇结构的Skyline查询算法,该算法利用优于树形路由结构的聚簇结构进行数据的存储,采用以数据为中心的数据存储策略,并优化了过滤方法。该算法分为节点过滤和数据元组过滤,在节点过滤阶段根据数据存储策略可以过滤掉大量无效的存储节点,减少了网络通信;节点过滤之后,将支配能力最强的数据元组作为过滤元组,该过滤元组可以最大程度上过滤掉有效节点中的无效数据元组,进一步减少通信开销。(2)提出了针对WSN的连续Skyline查询算法,该算法分为两个阶段:预计算阶段和Skyline数据维护阶段。其中,预计算阶段又分为数据集划分和数据存储两个阶段。在数据集划分阶段,将传感器节点第一次产生的数据传送至汇聚节点,并对数据进行分层;数据存储阶段根据数据划分策略,将数据集S划分成n个集合(1,2,..,9)9)),接着,将这n个集合分发到其它网内存储节点上。Skyline数据维护阶段就是根据n个集合对查询结果集进行有效地更新维护。(3)进行了大量的仿真实验,实验结果显示本文提出的Skyline查询算法在确保能耗的基础上仍然具有很好的性能。
[Abstract]:Wireless Sensor Network (WSNN: wireless Sensor Network) is a distributed multi-hop ad hoc network. The existing Skyline query algorithms based on single server can not be well applied to distributed multi-hop ad hoc networks such as WSN. It is of great theoretical and practical value to study the efficient and energy-saving Skyline query method based on WSN. According to the characteristics of data flow in WSN, the algorithms of snapshot Skyline query and continuous Skyline query processing based on data stream are studied in this paper. The main research work in this paper is as follows: 1) the Skyline query algorithm based on clustering structure in WSN is proposed. The algorithm uses clustering structure which is superior to tree routing structure to store data, adopts data-centric data storage strategy, and optimizes the filtering method. The algorithm is divided into node filtering and data element filtering. In the node filtering phase, a large number of invalid storage nodes can be filtered according to the data storage strategy, and network communication is reduced. After the node filters, the most dominant data tuples are taken as filtering tuples. This filter tuple can filter out invalid data tuple in the valid node to the maximum extent, and further reduce the communication overhead. (2) A continuous Skyline query algorithm for WSN is proposed. The algorithm is divided into two stages: the prediction phase and the Skyline data maintenance stage, in which the prediction phase is divided into two stages: data set partition and data storage. The data generated by sensor nodes for the first time is transmitted to the convergence node, and the data is stratified. In the data storage stage, the data set S is divided into n sets according to the strategy of data partitioning, and then the data set S is divided into n sets. During the data maintenance phase of distributing these n sets to other storage nodes in the network, the query result set is effectively updated and maintained according to n sets. A large number of simulation experiments are carried out. Experimental results show that the proposed Skyline query algorithm still has good performance on the basis of ensuring energy consumption.
【学位授予单位】：天津理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【相似文献】