Research On Data Aggregation Key Technology Of Wireless Sensor Network

Wireless sensor networks which sonsists of a large number of micro-sensor nodes are deployed in the monitoring region. Nodes in the area form a multi-hop self-organizing network system through wireless communication. This system can collect data from monitored region and transmit data to the client by the sink node, so as to achieving the target of monitoring interesting area. Due to the limited energy of wireless sensor nodes, the problems of how to reduce energy consumption and improve the lifetime of the network have become a hot research. Studies have shown that data aggregation algorithm has great impact on network performance. The design of effective data aggregation algorithm can prolong the network life.Based on the in-depth research of data aggregation algorithm, this thesis analyzes the strengths and weaknesses of the existing algorithms in wireless sensor network. For the problem of unbalanced energy consumption, the aggregation algorithms of the wireless sensor networks based on static nodes and mobile sink have been studied:(1) For the stationary wireless sensor networks, an energy efficient data aggregation routing algorithm based on clustering LEACH-EM is proposed. In the election of cluster head stage, it considers energy factor and network topology as a basis to select the cluster head. The threshold formula is added a constraint, so that the elected cluster head is more reasonable and after several rounds the threshold won’t be too low. In the cluster formation stage, the way of how the nodes join to a cluster is analyzed. Node chooses a cluster head which makes network communication with minimal energy to join. In the data transmitting stage, it proposes a method of forming a data gathering tree whose weights are improved, which can balance the energy consumption through multiple hops.(2) For the wireless sensor networks based on mobile sink, a data aggregation algorithm based on rendezvous point is proposed, namely DCRP. The algorithm can be divided into three stages:RP (Rendezvous Point) nodes selection, data aggregation within the grid and mobile sink path optimization. Since the position of RP nodes directly affects the path of mobile sink, in the stage of RP nodes selection, network is divided into grids and RP nodes is selected based on energy and network topology, so that the elected RP nodes can be closer and the number is reasonable. In the data aggregation stage, for the characteristics of the RP nodes’ positon, a secondary cluster structure is designed which can reduce the energy consumption of the nodes in remote area. At the last stage, simulated annealing algorithm is used to calculate the optimal path of RP nodes. Then the TSP problem is transformed into a TSPN problem which further shortens the time of data collection of mobile sink.Simulation results show that for stationary wireless sensor networks, the proposed algorithm can prolong the network lifetime. The algorithm which is used for mobile sink-based wireless sensor networks can effectively balance the network energy consumption and shorten the network delay.