Abstract: The mining technique of moving objects trajectory data is one of the important research branches of moving objects databases, and information derived from trajectory data can be applied to traffic control, climate forecast, ecological study and so on. A formal definition of temporal distance based on the trajectory data is provided firstly,and in order to decreasing the amount of computation in the process of query, this paper uses a relatively simple approximate distance to replace accurate distance. Distance query,instant nearest neighbor query and interval nearest neighbor query based on temporal distance arc implemented on relational model and tree structure. I}he query efficiency of two models is also compared, and the experimental results show that tree structure has better efficiency. In order to further improve the efficiency, this paper adopts embedded index in the tree structure model.

Key words: Moving objects, Temporal distance, Nearest neighbor qucry,Tree structure, Embedded index