Dual Grid: A New Approach for Robust Spatial Algebra Implementation
- 67 Downloads
Systems of spatial data types and operations, or spatial algebras, are fundamental for the implementation of spatial database systems. Several designs of such algebras have been proposed in the last decade, and recently commercial DBMS offer such algebras in the form of extension packages (e.g., “data blades”). However, actual implementations are generally severely restricted when compared to designs in the literature. A main reason is that at the implementation level one cannot further ignore the problems of robustness and topological correctness arising from the discrete number representations used in computers. Therefore, implemented packages either avoid “problematic” operations, or accept inconsistencies and topological errors in the answers of queries due to rounding effects.
The ROSE algebra , proposed and implemented earlier, goes a long way towards avoiding such problems, since it was defined from scratch with robustness problems in mind. It is founded on a discrete geometric basis called a realm. The ROSE algebra guarantees a correct behavior of operations and has an entirely robust implementation. Unfortunately, the realm concept and its interaction with DBMS are difficult to implement, and certain kinds of topological problems still remain.
In this paper we introduce the concept of dual grid, which provides a new approach for the representation of spatial information that avoids any robustness and topological correctness problem and allows a less restrictive implementation of spatial algebras. As an example, we show how can it be used for implementing the ROSE algebra without realms, and show that such an implementation does not suffer from the side effects and disadvantages of the original realm-based one.
Unable to display preview. Download preview PDF.
- 1.J. Bentley and T. Ottmann. “Algorithms for reporting and counting geometric intersections,” IEEE Transactions on Computers, Vol. C-28:643–647, 1979.Google Scholar
- 2.J. Davis. “IBM's DB2 spatial extender: Managing geo-spatial information within the DBMS,” Technical report, IBM Corporation, 1998.Google Scholar
- 3.D. Dobkin and D. Silver. “Applied computational geometry: towards robust solutions of basic problems,” Journal of Computer and System Sciences, Vol. 40:70–87, 1990.Google Scholar
- 4.M. Egenhofer. “Spatial SQL: a query and presentation language,” IEEE Transactions on Knowledge and Data Engineering, Vol. 6(1):86–95, 1994.Google Scholar
- 5.A. Fernandes, A. Dinn, N. Paton, M. Williams, and O. Liew. “Extending a deductive object-oriented database system with spatial data handling facilities,” Information and Software Technology, Vol. 41(1):483–497, 1999.Google Scholar
- 6.A. Forrest. “Computational geometry in practice,” in Fundamental Algorithms for Computer Graphics, Springer-Verlag, 707–723, 1985.Google Scholar
- 7.D. Greene and F. Yao. “Finite-resolution computational geometry,” in Proc. 27th IEEE Symposium on Foundations of Computer Science, 143–152, 1986.Google Scholar
- 8.L. Guibas and D. Marimont. “Rounding arrangements dynamically,” in 11th Annual Symposium on Computational Geometry, 190–199, 1995.Google Scholar
- 9.R.H. Güting. “Geo-relational algebra: a model and query language for geometric database systems,” in Proc. of the Intl. Conf. on Extending Database Technology, Venice, Italy, 506–527, 1988.Google Scholar
- 10.R.H. Güting, T. de Ridder, and M. Schneider. “Implementation of the ROSE algebra: efficient algorithms for realm-based spatial data types,” in Proc. of the 4th Intl. Symposium on Large Spatial Databases, Portland, Maine, 216–239, 1995.Google Scholar
- 11.R.H. Güting and M. Schneider. “Realms: a foundation for spatial data types in database systems,” in Proc. of the 3rd. Intl. Symposium on Large Spatial Databases, Singapore, 14–35, 1993.Google Scholar
- 12.R.H. Güting and M. Schneider. “Realm-based spatial data types: the ROSE algebra,” VLDB Journal, Vol. 4(2):100–143, 1995.Google Scholar
- 13.Ill94: 194, “Illustra 2D spatial datablade (Release 1.3) guide”.Google Scholar
- 14.P. Kornerup and D. Matula. “Finite precision rational arithmetic: an arithmetic unit,” IEEE Transactions on Computers, Vol. C-28:378–388, 1983.Google Scholar
- 15.V. Milenkovic. “Double precision geometry: a general technique for calculating line and segment intersections using rounded arithmetic,” in 30th Annual Symposium on Foundations of Computer Science, 500–505, 1989.Google Scholar
- 16.V. Muller, N. Paton, A. Fernandes, A. Dinn, and M. Williams. “Virtual realms: an efficient implementation strategy for finite resolution spatial data types,” in Proc. of the 7th Intl. Symposium on Spatial Data Handling—SDH'96, Vol. 2. Delft, The Netherlands, 11B.1–11B.13, 1996.Google Scholar
- 17.Ora97: 1997, “Oracle8i spatial cartridge”.Google Scholar
- 18.Ora99: 1999, “Oracle8i spatial: features overview”.Google Scholar
- 19.T. Ottmann, G. Thiemt, and C. Ullrich. “Numerical stability of geometric algorithms,” in 3rd ACM Symposium on Computational Geometry, 119–125, 1987.Google Scholar
- 20.F. Preparata and M. Shamos. Computational Geometry, Springer-Verlag: Berlin; Heidelberg; New York, 1985.Google Scholar
- 21.P. Schorn. “Degeneracy in geometric computation and the perturbation approach,” The Computer Journal, Vol. 37(1), 1994.Google Scholar
- 22.P. Svensson and Z. Huang. “Geo-sal: a query language for spatial data analysis,” in Proceedings of the 2nd Intl. Symposium on Large Spatial Databases, Zürich, Switzerland, 119–140, 1991.Google Scholar