Research On The Geometry-based Distributed Skyline Query Method In WSN

In recent years, environmental pollution and the quality problem of the air get the increasing attention. According to the actual situation, to realize the value of the wireless sensor network gets more and more concern. The skyline query strategies in wireless sensor networks have been widely used in the field of environmental detection, and these areas mostly are related to the spatial distance problem. Since the multi-dimensional of the spatial position, it brings the enormous computational cost for the general skyline query strategy with the aspect of the attribute calculation. In order to improve the efficient use of sensor energy, it needs a new query strategy to solve the skyline query on the spatial distance. This paper proposes a method of the geometry-based distributed skyline query(GDSky). The specific work is as follows.This paper uses the method of the geometry-based region partition to divide the whole query region into several sub-regions, so that it can determine the dominance relationship between sub-regions. This paper proposes a method of cut of the skyline area based on the convex hull vertices, which can quickly find the spatial skyline data with respect of a specific query area, and reduce the comparison times between nodes. This paper adopts the regional partition based on the triangulation method which divides the query region into several sub-regions. So it can carry out a distributed query in each sub-region. The paper also designs the clustering strategy in one sub-region, which can cluster on the data within a sub-region in order to query the skyline in parallel and save time. Besides, in order to traverse all the neighbor sensor detective locations, and cut the nodes which are dominated on the spatial distance. the paper also proposes the distributed query strategy based on the data node tree. To carry out the general skyline query on the non-spatial properties of the obtained spatial skyline,the paper also puts forward the non-spatial skyline query method. At the same time, the strategy conducts the spatial skyline query on the remaining spatial skyline data so as to realize the parallel execution. This paper presents the cut strategy among the sub-regional, which can cut all dominated nodes in one sub-region. And also puts forward the cut strategy in a sub-region, which can cut some dominated nodes in a sub-region and reduce the data communication in the network.Simulation results show that the method GDSky proposed in this paper can quickly find the places which are near to the query locations and have the larger pollution energy. The method also can efficiently reduce the times of comparison between data, improve the query efficiency, and reduce the energy consumption of sensor node.