Researches On Data Fusion Algorithm In WSN Based On Routing Technology

Wireless sensor network (WSN) data fusion technology is closely related to datacommunication in the network, so the combination with routing technology of data fusionalgorithm study is an important research direction in data fusion technique of WSN. The datafusion algorithm of WSN based on routing technology is mainly divided into planar fusionalgorithm and hierarchical fusion algorithm. This paper presents two novel improvedalgorithms based on LEACH algorithm which is the classic hierarchical fusion algorithm inWSN.To solve “hotspots” problem in the area far from the base station which caused byuneven energy consumption of the cluster nodes in LEACH algorithm, multi-hop modeimproved the communication way between fusion nodes and base station can be used. But itwill lead to a new "hotspots" problem in the near base station area, because the fusion nodesnear the base station need consume more energy to transmit the data form the nodes whichare far from the base station. In order to solve the above two problems, this paper uses thethought of non-uniformity to improve the LEACH algorithm and presents a novel datafusion algorithm base on non-uniform distribution double cluster head (DNDC). DNDCalgorithm use multi-hop way to transmit data that can resolve the "hotspots" problem in thearea far from the base station. And by using the non-uniformity the distance between thecluster and base station is closer, and the probability of selecting an assistant fusion node inhe cluster is greater. It can resolve the "hotspots" problem in the near base station area,because the fusion node in the near base station area does not need to consume excessiveenergy to forwarding the data which is from the node far from the base station.In order to further optimize the multi-hop routing path of the DNDC algorithm, thispaper applies ant colony optimization algorithm (ACO) to make an improvement on theDNDC algorithm, and presents a novel improved data fusion algorithm of DNDC based onAnt Colony Optimization (ADNDC). The traditional ant colony algorithm need a lot of antsand make many times of iteration to find the shortest path in the network, but this paper is aimto find the path of minimum energy consumption instead of shortest distance. In addition, alarge number of iterative calculations will consume a lot of energy. Therefore, traditional antcolony algorithm does not apply to the hierarchical fusion algorithm, so this paper firstly make presents a novel ant colony algorithm at one iterative calculation, which can find out allpaths in the network. Then, fusion nodes in the network select the next hop node for datatransmitting, according to the probability transfer formula. After the optimization of ACO,compared to the DNDC algorithm, ADNDC algorithm can make the energy consumptionminimum for transmitting data of the path from fusion node to base station.According to the topic design two algorithms, two different experiments areimplemented. Results show that the DNDC algorithm LEACH algorithm has a better networkenergy balance and longer survival time of the network. Also, the DNDC algorithm can adaptto different network sizes. While the ADNDC algorithm maintains the DNDC algorithm agood network energy balance based on DNDC algorithm, compared with a longer survivaltime of the network. Furthermore, it is more suitable for large-scale network structure.