Research Of Improvement Of Algorithms For Clustering In Wireless Sensor Networks

The nodes in wireless sensor networks are embedded devices with small-size and limited-energy. Its capacity of communication, calculation and storage is very limited. Well topology control can improve the efficiency of MAC protocol and routing protocol; It also can provide support for wireless sensor networks’data integration, communication and time synchronization; Balance the energy consumption of whole network and prolong the network’s life time. So the topology control is a crucial technology in wireless sensor networks. Node’s power control and cluster topology control are two research fields in topology control. Cluster topology control, also called clustering algorithm, in which several nodes are chosen as backbone node building a backbone network used for data integration and data forwarding.Improved algorithms are given in this thesis:(1) An improved clustering algorithm of LEACH is offered. A new threshold expression of cluster head option of LEACH in wireless sensor networks is given, which is based on residual energy, distance between a node and the sink node, and the times as cluster head. A new evaluation metrics based on residual energy, distance between the node and the cluster head node is used for cluster chosen. The simulation results show that the new arithmetic makes a great progress in energy consumption balance and network’s lifetime. Then the network’s life of new algorithm is twice over the LEACH.(2) A clustering algorithm method is brought forward based on dividing distance and residual energy, to make cluster head distribution more reasonable. Divide the networks in three districts by the distance of a node and the sink node. The node whose residual energy is maximum in a district is chosen as Cluster head. And when out of the Cluster head communication scope two nodes are chosen as Cluster heads by residual energy. Then the number of existing nodes is mach more than LEACH by simulation.