The Research Of The TEEN Protocol Of Data Fusion In Wireless Sensor Network

In wireless sensor networks, traditional data transfer method is to data collectedby each sensor node are sent to the sink-node. Because of the overlap of the sensornodes and intensive monitoring of the distribution of the region, will inevitably leadto the network has a lot of redundant data. Redundant data will cause networkdelayed, channel congestion, energy consumption increase, etc., for which we haveintroduced a data fusion technology in wireless sensor networks.Data fusion technology, is the number of data fusion and the application in thewireless sensor networks is the intermediate node from the source node receives thenumerous data according to user requirements analysis process fusion, will therepetition, low credibility or invalid data, eventually merge into a single highcredibility, high precision of data information, and then send it to the node. Based onthis, this article on the wireless sensor networks (WSN), as well as some existing datafusion algorithm based on wireless sensor networks routing protocol, on the basis ofanalysis and research to TEEN protocol as the basic model of hierarchical routingprotocol, proposed an improved stepwise data fusion algorithm and the data fusionmodel was established.The main idea of the stepwise data fusion algorithm proposed in this paper is: Inwireless sensor networks, the transmission process of the sensor nodes collected datato the sink-node is divided into three steps, according to the different characteristicsof each ladder receives transmitted data, adopting different data fusion algorithm. Thefirst step of the algorithm for cluster member nodes within the collection of datafusion, is the arithmetic average of the data fusion method which can reduce theamount of communication data from the data source; in the process of cluster membernodes send converged data to the cluster head, quoting the mechanism that based onthreshold interval data inspection, can effectively improve the transfer efficiency.Considering the different levels of confidence and accuracy of the member nodescollected, in the second step，using adaptive weighted fusion algorithm calculate thedata of the cluster head node obtained from cluster member nodes, which hascombined each member node data differently. In the third step, the cluster head node sends converged data to sink-node. Due to the sink-node set away from themonitoring area commonly, the transmission process will consume more energy,using the threshold value in the process of the control mechanism, in order to achievethe purpose that eliminate invalid and repeated converged data.Through the above three steps of data fusion, will be able to eliminate redundantdata in the network effectively, reduce the channel congestion and save energyconsumption of nodes, so as to prolong the life cycle of wireless sensor network.Finally, through the use of the NS-2simulation analysis confirmed the superiority ofthe stepwise data fusion algorithm.