Affiliations: Department of Computer & Information Science & Engineering, University of Florida, Gainesville, FL 32611, USA. gv1@cise.ufl.edu; mschneid@cise.ufl.edu
Note: [] Address for correspondence: E301 CSE Building, University of Florida, P.O. Box 116120, Gainesville, FL 32611-6120, USA
Note: [] This work was partially supported by the National Aeronautics and Space Administration (NASA) under the grant number NASA-AIST-08-0081.
Abstract: Data warehouses help to store and analyze large multidimensional datasets and provide enterprise decision support. With an increased availability of spatial data in recent years, several new strategies have been proposed to enable their integration into data warehouses and and perform complex OLAP analysis. Cardinal directions have turned out to be very important qualitative spatial relations due to their numerous applications in spatial wayfinding, GIS, qualitative spatial reasoning and in domains such as cognitive sciences, AI and robotics. They are frequently used as selection and restriction criteria in spatial queries. In data warehouses, cardinal directions can be used to perform spatial OLAP and feature navigation operations. In this article, we introduce and develop the Objects Interaction Graticule (OIG) approach to query the cardinal direction relations among spatio-temporal objects in data warehouses. First, we apply a tiling strategy that determines the zones belonging to the nine cardinal directions of each spatial object at a particular time and intersects them. This leads to a collection of grids over time called the Objects Interaction Graticule (OIG). For each grid cell, the information about the spatial objects that intersect it is stored in a Objects Interaction Matrix. In the second phase, an interpretation method is applied to these matrices to determine the cardinal direction between the moving objects. The results obtained for each valid instant over the objects' lifetime describe the variation in the objects movement over time. This is integrated as a spatio-temporal OLAP operation in a novel moving objects data warehouse (MODW) that provides an extensible framework for supporting complex structured objects. Finally, we define new directional predicates that extend MDX querying and leverage OLAP between moving objects.
Keywords: Cardinal directions, data warehouse design, spatio-temporal OLAP
