文档介绍：Abstract II AbstractThe primary function of wireless work (WSN) is to monitor the information in the region in an effective way, which requires every sensor could transfer the data to the cluster node through mediators. In order to make the data transfer more conveniently and save costs, the optimization of transmission path and the fusion of transmission data are needed. In this paper, the main research is data fusion technology in wireless work. Two improved algorithms of data fusion are presented for different kinds of data work structure, based on cluster routing. 1. A new method of cluster-based data fusion in work is presented. Firstly areas which are covered by wireless work are divided into many clusters by k-means algorithm and the cluster head node of every region cluster is chosen. Then a data fusion tree is constructed based on cluster head node in every region cluster. Through data fusion tree, cluster head node can collect information of all nodes in its region, which will help to produce the representative data of this region cluster. After that, data of all head nodes will be collected and then the data of the monitor region will be gotten from them. 2. A new based-on attribute-cluster spatio-temporal data aggregation query algorithm is proposed. Data aggregation uses clusters as WSNB-TREE’s storage object. Areas which are covered by wireless work are divided into attribute-clusters. The data of the homogeneity sensors are easy to do real-time statistics and segmental query. Areas were confirmed in advance are divided into a few data aggregation clusters,then the data of the heterogeneity sensors are aggregated. Hence, less time is taken to filter out the events which are needed to be queried, and then do spatio–temporal data aggregation queries in the work. Key words: wireless work,data fusion,clustering, data aggregation, WSNB-TREE 学位论文独创性声明本人郑重声明: 1、坚持以“求实、创新”的科学精神从事研究工作。 2、本论文是我个人在导师指导下进行的研究工作和取得的研究成果。 3、本论文中除引文外,所有实验、数据和有关材料均是真实的。 4、本论文