| Wireless Sensor Network(WSN)is a huge research system,it is a combination of information technology and electronic technology,including wireless transmission technology,topology technology,embedded technology and so on.How to extend the working life of WSN has been a hot topic of research.There are also many ways to improve the working time of the network,such as optimizing the topology of the network,using energy replenishment,optimizing the routing protocol of the network,and so on.With the development of microelectronics technology,energy harvesting WSN(EH-WSN)based on environmental energy supply is favored by researchers because of its network working life is significantly longer.At present,domestic and foreign scholars have done a lot of research on EH-WSN,but there are still some shortcomings.In the research process of EH-WSN clustering algorithm,there is no clear requirement for the number of network clusters;the selection of cluster head is not comprehensive enough.Therefore,this dissertation proposes a self-powered wireless sensor network clustering algorithm based on fuzzy control.After simulation experiments,it proves the superiority of the proposed algorithm compared with the traditional algorithm.The results of the dissertation are as follows:(1)Research on self-powered clustering algorithm based on fuzzy logic.First of all,in view of the existing self-powered wireless sensor network clustering algorithm,which does not consider the optimal number of clusters in the network,the solar replenishment model is introduced into the network energy consumption model.And the function relationship between the total energy consumption of the network and the number of network clusters was obtained for each round.The function was derived to obtain the optimal number of clusters in the network.Secondly,to solve the problem of unbalanced energy consumption in the network,that is,the problem of cluster head selection when clustering,this dissertation uses a two level fuzzy decision system to evaluate whether a node in the network can become a cluster head node.First,the remaining energy of the node and the number of neighboring nodes are input into the first level(capability level)as the judgment index.All nodes are screened by the Mamdani fuzzy control system.The node that is required by the threshold value becomes the candidate cluster head node;then the node’s centrality parameter and the proximity parameter are input into the second level(cooperation level)as the judgment index to perform secondary screening on the candidate cluster head nodes with the same control system,and finally obtain the optimal network cluster head node under the comprehensive reference.(2)Using MATLAB simulation tools,the proposed algorithm and similar algorithms are simulated and compared under the same basic parameters.First of all,in the case of solar power supply,the proposed algorithm in this dissertation is significantly better than the classic algorithm in terms of the number of surviving nodes in the network,and then the network energy consumption of the proposed algorithm is similar to the network energy consumption of the comparison algorithm.The network data throughput of the algorithm is obviously better than that of the comparison algorithm.At the same time,in order to improve the general applicability of the algorithm,the performance and working life of the algorithm in this dissertation are tested in the case of no energy supply.The simulation results show that the algorithm in this dissertation still has advantages in the number of live nodes in the network when there is no energy supply compared with the comparison algorithm,and after the algorithm optimizes the network routing,the proposed algorithm can reduce the network data without high energy consumption.Packet throughput is significantly better than other comparison algorithms.Simulation experiments show that,regardless of whether the proposed algorithm has energy supply or not,the proposed algorithm prolongs the working life of the network,and the network data throughput is also improved when the network energy consumption is low. |
