Modeling and Querying Interpolated Spatial Data

We consider the modeling of complex spatial data that require the interpolation of values from collections of samples. We rst show that the approach based on extending DBMSs with new types is too restrictive to meet the needs of data such as moving objects, elevation maps, or other interpolated spatial data. We propose a data model based on the constraint database paradigm, which ooers major improvements on the state-of-the-art solutions. It provides an elegant and very simple logical data model, based on innnite sets which can be manipulated through their deenition with constraints. This model encompasses classical maps as well as complex interpolated data. Finally, query languages can be deened by classical means, based on relational calculus or algebra, with a low data complexity, suitable for most queries in practice. It is shown in particular that the complexity of the queries which depends generally exponentially upon the dimension of the data, can be reduced drastically.