Research On Clustering And Energy-saving Algorithm Of Wireless Sensor Network

Wireless sensor network, which is called WSN for short, integrates sensor technology, embedded computing technology, distributed information processing technology and wireless communication technology. It is a new field of computer science and technology, and has a very broad prospect, which has attracted much attention of the academia, the military sector and the industrial sector.In order to reduce energy consumptions of sensor nodes, provide a flexible and reliable communication, and increase the network's scalability, researchers usually use a network hierarchical structure in WSN. The studies on the hierarchical structure of WSN dig into solving two major problems: first, how to divide clusters and work mode of nodes, second, how to construct the route to the base station. Taking these two aspects as the starting point, the thesis, based on LEACH and combining the research achievements from home and abroad, studies and explores the hierarchical routing algorithm of WSN and proposes a clustering and energy-saving routing algorithm based on the cluster domain, which is called ECD algorithm for short. The method of dividing areas, the best number of the cluster heads, the clustering method, and how to build a routing tree are studied in the thesis.The nearer the cluster head is to the base station, the more possible it communicates with the base station directly, the faster its energy consumes. The thesis hives off an area close to the base station, where cluster heads communicates with the base station directly and shares the task of communication with the base station, in order to balance the energy consumption of cluster heads, thus extend the network's survival time. The number of clusters of WSN will affect the energy consumption rate of the network. The thesis discusses how to identify the best number of clusters on condition that the overall energy consumption is the smallest. A reasonable clustering scheme can save and balance the overall energy consumption when gathering the data. The thesis proposes a clustering algorithm, which elects the node with the highest energy within the cluster to be the cluster head, ensure that the cluster head in the central cluster, and make cluster heads distribute well. The location and energy of the cluster head must be considered at the same time when constructing a routing tree, and the consider how to choose the path. The thesis proposes a scheme of how to construct a routing tree.Simulation shows that the algorithm survives a long time, the load is balanced to a large extent, and the complexity is low. Furthermore, the survival time of the network grows slow with the increase of the distance from the base station to the monitoring region.