DIME: Disposable Index for Moving Objects

Increasing usage of location-aware devices, such as GPS and RFID, has made moving object management an important task. Existing spatial-temporal indexing techniques support efficient queries on large number of moving objects. In these techniques, significant I/O is consumed by removing obsolete locations, which impairs the performance of moving object management. On the other hand, some techniques have been designed to index moving objects in main memory to facilitate frequent location updates. However, they are limited by the size of available memory. In this paper, we propose a generic spatial-temporal index framework, Disposable Index for Moving objects (DIME), to efficiently handle location management over mobile agents with hybrid storage support. The proposed disposable index framework eliminates delete operations on the spatial indexing structure and processes insert operations in memory only. Most existing spatial indexing structures can be adopted in this generic framework. Both snapshot and continuous query processing has been designed for this framework. Experimental results on benchmark data sets demonstrated the scalability and efficiency of DIME.