Research On Hybrid Topology Controlling Algorithm In Wireless Sensor Network

Wireless Sensor Network(WSN)is a novel form of network. The nodes of WSN have functions of sensing, information processing and wireless telecommunication. These functions make WSN great value and widely used in practice among the fields such as military affairs, environment, medical treatment, in house and other commerce business and WSN is a hotspot problem in scientific filed. In traditional wireless network, the main purpose of network protocol design is to provide high QoS and efficient utilization of bandwidth, but the first priority task for routing protocol design in wireless sensor network is to increase the energy efficiency of sensor nodes and prolong lifetime because of limited energy, communication and computation capability of sensor nodes. Because of the difference between WSN and traditional network mentioned before,research of wireless sensor network will propose many new challenges to network researchers.Topology control is an important filed in WSN research. An excellent topology controlling technology can decrease the power consume and increase the throughput of the network. The hierarchical topology controlling algorithm can be designated as clustering algorithm. It chooses heads for each cluster and organizes these heads to be a core network that can process data and relay packets. And the common member nodes can power down most of the time in order to save energy.In this paper, we introduce the research, characteristics and typical topology controlling protocol of WSN at the beginning. And then the classical clustering algorithm-LEACH is analyzed. Based on it, Hybrid Topology Controlling algorithm based on Subtractive Clustering(HTCSC) is proposed. It introduces the least mean squared subtractive clustering to divide clusters and elect the initial heads which can make the network topology more reasonable. A head changing strategy based on nodes'energy estimating and cluster architecture is also proposed, and it can prevent that the nodes with low energy is elected to be the head so that the cluster is rapidly jammed. It also issues a shortest path routing algorithm based on energy cost and it can decrease the power consume via long distance transmitting. HTCSC just considers on how to decrease the energy consume in each cluster, but doesn't balance it between the clusters. So, a Hybrid Nonuniform Clustering Topology Controlling algorithm(HNCTC) for balancing is proposed. Considering the density, energy and the location of sink node in the network, a nonuniform clustering mechanism is proposed which can balance the energy load among the cluster heads. A head changing strategy based on energy estimating is also proposed, and it can prevent the nodes dying too early while it is able to maintain high energy efficiency.Simulation result shows that HTCSC and HNCTC are able to implement the advanced topology. They can decrease the total energy and balance the power consume of each node. While they are both able to prolong the effective system life and can increase the effective packets.