Trajectory processing and analyzing is now one of the hottest research issues in related fields, such as geography, database, and so on. Inspired from the Stop/Move abstract model of trajectory, this paper aims at modeling trajectory by integrating context geographical information and studying the processing techniques on trajectory spatio-temporal pattern queries. First, the semantic relationships between Stop/Move objects and point/line/polygon features are analyzed, based on which a novel trajectory model that explicitly expresses geographical information associated semantics is proposed; Next, an application-independent trajectory relational schema for this trajectory model is designed; Next, the concept of trajectory spatio-temporal pattern query is introduced, and its SQL processing framework, based on trajectory relational schema, is also proposed. Finally, this paper discusses how to processing trajectory spatio-temporal pattern queries with pure SQL languages. To make the answering procedure more clearly, this paper not only presents several typical example queries and their corresponding SQL statements, covering all three types of trajectory spatio-temporal pattern queries, i.e., location-time, location-order and location-relation, but also analyze in detail the hidden SQL processing steps. The trajectory model and its sptaio-temporal pattern querying, proposed in this paper, is a relatively new solution to process and analyze trajectory dataset. It is not only capable to model trajectory and its context geographical information, but also a cost-efficient way to process trajectory spatio-temporal pattern queries, which is carried out just based on mature SQL technology without the necessary to develop complicated data mining methods.
