Research Of Data Fusion Algorithm Based On The Internet Of Things

In today’s society, various of communication and technology update unceasingly,at the same time, the development of network has changed from the Internet to thefield that all of the machines can communicate to each other, and achieve theNetwork Everything goal, which is called the network communication. The Internetof things (Internet of Things, IOT) can be called as the important part of the newgeneration of information technology, it does realize the intercommunication betweenthings and things, things and man, man and man.In the Internet of Things, the node distribution is very wide, and the number isvery huge. In this situation, it is very important to fuse the data in the perceptual layer.Data fusion technology has been developed very mature, but compared with othernetworks the Internet of things has its own characteristics. So when the data fusionalgorithm is directly used in the Internet of things, there will be many problems. Thispaper will analysis these problems and find the concrete improvement program.One of the characteristics of Internet of things is the mass of data. If the datafusion algorithm is directly used in the Internet of things, it may cause a lot energyconsumption of the fusion node, and may lead the node’s death. So we need toimproved the data fusion algorithm to make it adapt to the the characteristics of theInternet of things. In addition, there are different real-time requirements for differentkinds of data, if the data is transmitted according to the original route, even if thecorrect fusion processing, the result is not what the Internet of things need which doesnot reflect the different real-time requirements. So before we improve the data fusionalgorithm we should first improve the transmission of data, lay the premise for theaccuracy of data fusion results, this can make the obtained result is what the Internetof things need.To accomplish the above tasks, this paper mainly works on two aspects. First, inorder to make the fused data to meet the characteristic that different kinds of data hasdifferent real-time requirement. this paper selects the LEACH algorithm as prototypeand improves it. The improved algorithm makes different kinds of data can be transmitted in different lengths of time, and makes sure that the fusion result can meetthe characteristics of the Internet of things, and lays a good foundation for data fusion.Then, according to the massive data in IOT, the paper chooses fuzzy algorithm asprototype, and sets the threshold to reduce the amount of data which needs to be fused,and introducing the concept of weight values to make it more suitable for theremaining data fusion, and finally makes the data fusion method suitable for the IOT.At the end of the paper, we do simulation to the improved algorithms which arementioned in this paper.The simulation results shows that the improvement of thedata fusion based on the characteristics of the Internet of Things reaches the expectedeffect, and the data fusion is also better adapted to the Internet of Things.