Given two sets
S1, S2 of moving objects, a future timestamp tq, and a distance threshold d, aspatio-temporal join
retrieves all pairs of objects that are within distance d at tq. The selectivityof a join equals the number of retrieved pairs divided by the cardinality of the Cartesian product
S
1กมS2. This paper develops a model for spatio-temporal join selectivity estimation based onrigorous probabilistic analysis, and reveals the factors that affect the selectivity. Initially, we
solve the problem for 1D (point and rectangle) objects whose location and velocities distribute
uniformly, and then extend the results to multi-dimensional spaces. Finally, we deal with nonuniform
distributions using a specialized spatio-temporal histogram. Extensive experiments
confirm that the proposed formulae are highly accurate (average error below 10%).