Nearest Neighbor Queries in a Mobile Environment

Nearest neighbor queries have received much interest in recent years due to their increased importance in advanced database applications. However, past work has addressed such queries in a static setting. In this paper we consider instead a dynamic setting where data objects move continuously. Such a mobile spatiotemporal environment is motivated by real life applications in traffic management, intelligent navigation and cellular communication systems. We consider two versions of nearest neighbor queries depending on whether the temporal predicate is a single time instant or an interval. For example: "find the closest object to a given object o after 10 minutes from now", or, "find the object that will be the closest to object o between 10 and 15 minutes from now". Since data objects move continuously it is inefficient to update the database about their position at each time instant. Instead our approach is to employ methods that store the motion function of each object and answer nearest neighbor queries by efficiently searching through these methods.