Research On Indoor Moving Objects Index Based On DR-tree

With the development of wireless communication technology and location-basedservices, data management of moving objects has become a research hotspot. In recentyears, scholars at home and abroad have made lots of researches on historical data indexof moving objects for various practical applications and proposed a series of indexschemes. Most of these schemes are based on outdoor space which support efficientspatial and temporal queries, however, they hardly provide efficient queries based onobjects due to the lack of object index. With the development of indoor positioningtechnology, indoor location-based services are becoming more and more properous, whilethe specialty of the indoor environment makes outdoor index schemes difficult to bedirectly applied to indoor space.A new historical data index scheme based on R-tree for the indoor environmentnamed DR-tree is proposed to manage the moving objects. The object dimension isindexed and treated equally with spatial and temporal dimensions, meanwhile, theDR-tree decouples the object and spacial dimensions to form a dual-tree structure. Inorder to refine the access methods, a window split algorithm is raised to split the originalsearch window. It not only supports efficient spatio-temporal query, but also providesefficient trajectory query based on objects.The DR-tree is realized in Windows operating system, using java language. With aset of test data, comparasion are made between DR-tree and RTR-tree to verify theperformance of DR-tree in spacial-temporal query and trajectory query based on objects.The results demonstrate that the DR-tree has high-efficient query performance.Finally, the DR-tree is put in use of a large-scale enterprise’s attendance system,according to the actual demands and condition of the system, employees’ attendancerecords are processed and indexed by DR-tree to provide highly effective queryprocessing.