The Nearest Neighbor Query In Mobile Object Database System

With the rapid development of technology, global positioning system and communication technology can record the positions of moving objects. The continuous movement of moving objects results in new challenges to database technology, and moving objects database appears.The optimizing of query is the focus problem in moving objects database, and the Nearest Neighbor Query is one of the most important operations. Traditional methods of the Nearest Neighbor Query all regard the query point as a static one,as the development of the technology in wireless communication and tracing moving objects, it becomes the new research focus and difficult point on how to extend the Nearest Neighbor Query method from static environment to moving objects environment. At the present time, researchers have done some work on the moving objects which orbits are straight lines, and proposed some methods of the Nearest Neighbor Query, such as the CNN algorithm. Compared with the traditional Nearest Neighbor Query, the Nearest Neighbor Query of moving objects almost regards the objects'position as a straight line, and it purpose to find one or more nearest neighbors for every static point on the moving straight line, so it's much more complex.This paper has studied the most popular index structures in spatial database, such as R-tree, TPR-tree etc., and analyzed their advantages and disadvantages. On that base this paper has introduced the current and popular algorithms of Nearest Neighbor Query, and analyzed their advantages and disadvantages, then given the scheme of improvement.Because the current algorithms of Nearest Neighbor Query almost are focused on processing the Euclidean distance of static objects and the moving objects which orbits are straight lines, this paper proposes the methods of the Nearest Neighbor Query in the road traffic network.