A Logical Formulation of Probabilistic Spatial Databases

  • Authors:
  • Austin Parker;V. S. Subrahmanian;John Grant

  • Affiliations:
  • -;-;-

  • Venue:
  • IEEE Transactions on Knowledge and Data Engineering
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

There are numerous applications where there isuncertainty over space and time. Examples of such uncertaintyarise in vehicle tracking systems where we are not always surewhere a vehicle is now (or may be in the future), cell and satellitephone applications where we are not sure exactly where a phonemay be, and so on. In this paper, we propose the concept of aSpatial PrObabilistic Temporal (SPOT ) database which containsstatements of the form "Object O is in spatial region R at sometime t with some probability in the interval [L, U]." We definethe syntax and a declarative semantics for SPOT databasesbased on a mix of logic and linear programming, as well asquery algebra. We show alternative implementations of some ofthese query algebra operators when the SPOT database has adisjointness property. Though the declarative semantics of SPOTdatabases is rooted in linear programming, we have found veryefficient algorithms that do not use linear programming methods.We report on experiments we have conducted that show that thesystem scales to large numbers of SPOT atoms as well as tofairly fine temporal and spatial granularity.