Predictive Query Processing Considering the Movement of Both User and Objects

Recently, with the increase in use of mobile devices such as smart phones and tablet PCs with GPS, it is possible to analyze a large volume of data aggregated from various sensors. Accordingly, a variety of location-based services (LBSs) have attracted attention. To effectively provide these services, techniques for efficient spatial query processing have been studied. In this paper, we propose a method to overcome the limitation of not returning the desired query result to the user, because existing studies did not consider movement of the user. Specifically, we propose an algorithm to efficiently process a predictive query in the road network that returns the best available K moving objects, in consideration of the time of the user's moving and that of the user's waiting. In this process, we apply the technique to gradually expand the range of user and object's movement simultaneously. Also, an appropriate index structure is used to efficiently process queries even in the road network with a large number of vertices and moving objects. Experimental results reveal the difference in the query result compared to existing studies and also reveal significant results in terms of efficiency.