Indexing

Abstract

Applications such as air traffic control and weather forecasting require the monitoring of a large number of moving objects; hence they require large spatio-temporal databases. Efficient query processing for large spatiotemporal databases is an important challenge.

For example, when monitoring thousands of airplanes and clouds, even simple-looking queries like “find all the airplanes that are currently traveling within some cloud” become difficult to evaluate in a naive approach that tries to find for each pair of airplane and cloud whether they intersect. Indexing is a way to efficiently evaluate such queries.

In this chapter we assume that the moving objects are represented as linear parametric rectangles, and we describe an indexing method for linear parametric rectangle relations. For the rest of this chapter parametric rectangle will always mean linear parametric rectangle.

Section 20.1 describes minimum bounding parametric rectangles, which are parametric rectangles that enclose a set of other parametric rectangles. This section presents an algorithm to find a minimum bounding parametric rectangle.