Research On Clustered-Network In Wireless Sensor Networks

The wireless self-organized sensor network, which is integration of sensor techniques, MEMS techniques and network communication techniques, is an innovative technique of information acquisition and processing. It is an advancing research field of multi-subject intersection. It brings wide application future in a good many fields, such as military, industry, medical treatment, transportation and commercial. A few achievements have been acquired in the world of recent years, but on some key techniques, there are also a lot of problems in need of resolution.This paper did a deep study in clustered-network created through hierarchical toplogy management, including optimization of energy consuming and logical toplogy in culustered-network, and construction of cluster-tree backbone network.In view of the energy-irreproducibility of the sensor network, how to improve the efficiency of energy consuming in sensor network and maximize its life time is one of key issues in wireless sensor network technology. On the basis of analyzing single-hop clustered-network and establishing network model and mathematics model, this paper discussed the relationship between several key network parameters and energy consuming,then found the optimized parameters that would achieve minimum energy consumption.The optimization of logical toplogy of clustered-network is a NP-hard problem. Based on the fundamental framework of genetic algorithm(GA), an improved algorithm is proposed in this paper, in which a new crossover, mutarion and selection method are used, The convergence process of genetic algorithm is accelerated by using the stratage of second selection. Simulations show that the global approximate optimal solution of the problem can be rapidly found.In order to solve the problem of inappropriate cluster head election and excessive long transmission distance, EACT ( Energy-Aware Cluster-Tree )algorithm is proposed to organize the cluster tree backbone network .Based on root tree and Maximal Weighted Independent Set, EACT utilizes periodical rotation of cluster heads to balance the energy consumed by nodes. Simulation results demonstrate that EACT would significantly increase the lifetime of the network..