Research On Clustering Topology Control Algorithm Based On Wireless Sensor Network

Wireless sensor network is a new wireless network in the 21 st century. Its appearance greatly affects people's life and work. As limited energy of sensor nodes and the adverse deployment environment, it is difficult to replace the battery in the actual operation. The problem of limited energy is always the key issue to limit the development of Wireless Sensor Network. In order to make maximize energy of nodes in network, it becomes central in the way of deploying nodes to use energy efficiently and extending the network lifetime in wireless sensor network.In this regard, this essay focuses on the topology control algorithm based on clustering and analyzes the deficiencies clustering algorithm in detail. Then an optimization of GAF Algorithm based on SVR and circular region division is proposed. Then, an optimization of LEACH algorithm based on SVR and PSO is proposed at the same time.The work in this essay is as follows:(1)For the lack of cells division of GAF algorithm and the problems of connection,an optimization of GAF Algorithm based on SVR and circular region division is proposed. The optimization of GAF algorithm divides into two parts. First of all, this essay imports SVR model to optimize the division of virtual cells and divides the original square areas into circular areas. Secondly, combined with the connectivity of the network, this essay changes communication radius of nodes. Finally, the results of the simulation experiment show that the proposed algorithm can optimize network performance, reduce energy consumption and extend network lifetime.(2)For the unreasonable distribution of cluster heads and the random selection of cluster heads of LEACH algorithm,an optimization of LEACH algorithm based on SVR and PSO is proposed. SVR is introduced into the proposed algorithm firstly.Considering the specific location of the nodes, supported vectors are searched and divide network so that each of region can produce cluster heads and enhance the problem of unreasonable distribution of cluster heads. Then this essay optimizes method of cluster heads selection so that the set of candidate cluster head nodes can be selected and become the initial particles of PSO. PSO searches the best cluster head by iterative methods. Finally, the results of the simulation experiment show that the proposed algorithm has certain advantage of reducing energy consumption and extending network lifespan.