Research Of Data Aggregation Algorithm Based On Spatial Correlation In Wireless Sensor Networks

Wireless Sensor Network(WSN)is a bran-new information acquiring and processing technology.The battery of sensor node in network is of one-off,which cannot be changed generally.Accordingly,energy consumption issue is one of the hot topics in WSN researches.Data aggregation technology can enhance network performance in data-aware,data processing,data transmission and so on,it becomes one of the core technologies in WSN.In practical applications,the sensor nodes are usually high-density deployment.How to identify redundant nodes effectively in the network,reducing redundancy and saving resources,has become an important research direction in the data aggregation.In this thesis,we analyze,expound and point out existing problems of research status and technical characters of sensor node model,data correlation and data aggregation algorithms in WSN.Directing at WSN of densely sensor nodes deployment,we propose a calculation methodology of redundant sensors based on grid-coverage model.Our method can efficiently recognize redundant nodes according to the degree of coverage of each grid.The boundary is included in recognition scope and computation complexity is reduced in our method.Combining spatial correlation and period-temporal correlation of data,we represent a novel data aggregation algorithm based on spatial correlation,which is one of the important innovations of the paper.The spatial correlation of nodes is determined by the location and the similarity of the data.It makes clustering results reasonable and accurate.Data period prediction can avoid the communication conflict between sensor nodes.Relay transmission of redundant nodes also can take advantage of network resources.We use IBRL dataset and LUCE dataset to complete simulation computation,comparing with other existing algorithms,we verify the superiority of our methodology in saving nodes' energy,reducing data transmission quantity and extending the network lifetime.