The Research Of Clustering Topology Based On Modified Particle Swarm Optimization

Wireless sensor network is a comprehensive frontier research field involving a wide range of disciplines.In view of the energy-constrained and multi-hop,routing protocol becomes the difficulty of WSN field.The clustering routing protocols is the core of WSN routing protocols because of its low power consumption and easy expansion.And as an important part of clustering routing,clustering topology technology is also paid much attention.In this paper,we firstly introduce a hierarchical clustering algorithm based on gradient with gateway(HCAGG).This algorithm considers the residual energy and the average distance to neighbor nodes to construct comprehensive weight in the cluster head election.At the same time,the gateway node effectively reduces the energy cost of the long-distance communication between the distant cluster heads.However,the algorithm does not consider the distribution of the energy in the cluster.In addition,the selection of gateway nodes is dependent deeply on the gradient.Aiming at the problems existing in HCAGG,a clustering topology algorithm based on modified particle swarm optimization(CTMPSO)is proposed.To solve problem of standard PSO,a modified particle swarm optimization algorithm(MPSO)which introduces the visual field and random behavior of the artificial fish swarm algorithm is proposed.MPSO can enhance diversity of particle and avoid PSO prematurely into the local optimum.Then verify the effectiveness of the MPSO algorithm by Sphere and Ratrigrin.At the same time,considering the characteristics of WSN and MPSO,establish a multi-food source foraging scene and a continuous evaluation function which are suitable for discrete WSN,so that MPSO can be better applied to WSN.In addition,considering the characteristics of the optimization target,comparative method is proposed to determine the search area.Aiming at the problem of HCAGG cluster head,the modified particle algorithm is used to search the best optimal node as the new cluster head which is taking residual energy,the average distance to neighbor nodes and the energy distribution ofthe nodes into account.Aiming at the problem of gateway,a two-hop optimal gateway is proposed and optimized by MPSO.The simulation results show that the CTMPSO algorithm can effectively prolong the lifetime of the network.Finally,the maintenance and updating strategy about on and out lines based on the CTMPSO algorithm are given to improve the adaptability of the algorithm.At the same time,according to the characteristics in the emerging field of WSN,the maintenance and updating strategy of nodes moving is also given to expand the application scenario.