The Research On The Algorithm Of The In-network Data Aggregation In Wireless Sensor Networks

Wireless sensor networks (WSN) is a kind of wireless self-organizing networks with a distinct data-centric property, which consists of a large number of sensor nodes that integrate sensor, the unit for data processing and the module for the short distance wireless communication. The objective of the sensor networks is to sense, collect and process the information of the various environments or the detected objects in the coverage area of the networks, and then send the sensing information to the user. Wireless sensor networks, together with utility computation, plastic electronics and body bionics, is considered as the four high-tech industries in the future around the world. As a new generation of the detecting and control system, the data gathering in sensor networks is the most fundamental and important function in the network system, and the performance of the data aggregation mechanism directly influences the energy conservation and the capability of the sensor networks.The research of the dissertation focuses on data aggregation of the in-network data processing in sensor networks. Based on the systematic analysis of the present data aggregation and in-network data processing mechanisms in sensor networks, the new data aggregation mechanisms to the situation of sparse data gathering are presented with the consideration of saving the energy resource and improve the network capability to the largest extent. And the mechanisms would perfect data aggregation research in sensor networks. The main work is as follows:First, the dissertation summarizes the data aggregation mechanisms in wireless sensor networks, and based on the model about the energy consumption in data aggregation, analyses the construction of the aggregation tree and the process of data aggregation in detail. The dissertation also does a theoretical analysis and comparison of energy consumption to three typical data aggregation mechanisms.Second, in order to achieve the data aggregation of the bit nodes, the dissertation puts forward a advanced algorithm of constructing data aggregation tree beginning from the perceptive node. The algorithm based on the probe and recall protocol, distributed construction one or more aggregation tree through comparing the nodes'weights. The theoretical analysis and the simulation test the validity of the algorithm.Third, to reduce the consumption of the energy, in the processing of constructing the aggregation tree,the process of constructing aggregation tree and data aggregation can be one using the weights of the nodes and the theory of probability. The advantage of the algorithm has been tested by theoretical analysis and the simulation.