Incremental processing of continual range queries over moving objects

Efficient processing of continual range queries over moving objects is critically important in providing location-aware services and applications. A set of continual range queries, each defining the geographical region of interest, can be periodically ( re) evaluated to locate moving objects that are currently within individual query boundaries. We study a new query indexing method, called CES-based indexing, for incremental processing of continual range queries over moving objects. A set of containment-encoded squares (CES) are predefined, each with a unique ID. CESs are virtual constructs ( VC) used to decompose query regions and to store indirectly precomputed search results. Compared with a prior VC-based approach, the number of VCs visited in a search operation is reduced from (4L(2) - 1)/3 to log(L) + 1, where L is the maximal side length of a VC. Search time is hence significantly lowered. Moreover, containment encoding among the CESs makes it easy to identify all those VCs that need not be visited during an incremental query ( re) evaluation. We study the performance of CES-based indexing and compare it with a prior VC-based approach.