Research On WSN Energy Efficiency Optimization Based On Improved Gray Wolf Optimization Algorithmimes

Wireless sensor network is a self-organizing network composed of a large number of sensor nodes with sensing,computing and communication capabilities,which realizes the organic fusion of logical information and the objective physical world.With the maturity of sensing technology,embedded computing technology and wireless communication technology,sensor technology has also been improved.Modern sensors have the advantages of low power consumption and multi-function,and can realize information collection,processing and The multiple functions of communication enable WSN to develop rapidly.At the same time,it has also received widespread attention as an important support for Internet of Things applications.Most of the WSN nodes are powered by batteries,and the energy is very limited.With the continuous development of the Internet of Things and 5G technology,the scale of WSN continues to expand,and a large amount of collected data needs to be processed in the cloud.Energy constraints have become the biggest problem restricting the indepth development of WSN.This thesis first briefly describes the research background and significance of WSN energy efficiency optimization,and then gives an overview of WSN and WSN energy consumption.Then,the relevant technologies of energy efficiency optimization in WSN are briefly described,among which,the basic idea and basic process of cluster routing are introduced emphatically.Finally,this thesis chooses to focus on WSN energy efficiency optimization from the perspective of clustering routing.In the three stages of clustering routing,cluster head selection and clustering,data fusion and inter-cluster routing respectively propose corresponding optimization methods to reduce Network energy consumption,improving the efficiency of data transmission,the main work of the thesis is summarized as follows:(1)Clustering and cluster head selection stage.Firstly,the relevant algorithms of clustering and cluster head selection are introduced,and then the gray wolf optimization algorithm used in this thesis is introduced.Aiming at the shortcoming that the gray wolf algorithm is easy to stagnate,by mixing the differential evolution algorithm and the gray wolf optimization algorithm,the convergence factor and scaling factor in the hybrid algorithm are improved.And considering the remaining energy of the node and the distance between the node and the sink node,the fitness function is designed to optimize the selection of the WSN cluster head node.(2)Intra-cluster data fusion stage.Firstly,the technical characteristics and related algorithms of data fusion in WSN are introduced.In order to reduce the redundancy of data and improve the accuracy of data fusion,this thesis adopts BP neural network as the method of data fusion.Aiming at the problem that the BP neural network algorithm is sensitive to the initial parameters during data fusion and is easy to fall into the local optimal solution,resulting in insufficient accuracy of the fused data,by fusing the cuckoo algorithm,designing a new nonlinear convergence factor,and using chaotic mapping to initialize the population The gray wolf optimization algorithm is improved by means of the improved gray wolf optimization algorithm,and then the initial weight and threshold of the BP neural network are optimized to improve the accuracy of the fusion data.(3)Inter-cluster routing stage.A discrete gray wolf optimization algorithm is designed for routing path optimization.Among them,the individual variablelength encoding method,fitness function and individual movement rules are designed,and the algorithm flow of routing path optimization based on the gray wolf optimization algorithm of variable-length individual encoding is elaborated to obtain the best multi-hop routing path.In this thesis,Matlab software is used for simulation experiments,and the optimization algorithms proposed in each stage are compared and analyzed in experiments.The results show that the algorithm proposed in this thesis has better performance in balancing network energy consumption,prolonging network life cycle and data fusion accuracy.