STEPQ: Spatio-Temporal Engine for Complex Pattern Queries

Dongqing Xiao, Mohamed Y. Eltabakh

Abstract: With the increasing complexity and wide diversity of spatio-temporal applications, the query processing requirements over spatio-temporal data go beyond the traditional query types, e.g., range, kNN, and aggregation queries along with their variants. Most applications require support for evaluating powerful spatio-temporal pattern queries (STPQs) that form higher-order correlations and compositions of sequences of events to infer real-world semantics of importance to the targeted application. STPQs can be supported by neither traditional spatio-temporal databases (STDBs) nor by modern complex-event-processing systems (CEP). While the former lack the expressiveness and processing capabilities for handling such complex sequence pattern queries, the later mostly focus on the Time dimension as the driving dimension, and hence lack the power of the special-purpose processing technologies established in STDBs over the past decades. In this paper, we propose an efficient and scalable spatio-temporal engine for complex pattern queries (STEPQ). STEPQ has several innovative features and ideas that will open the research in the area of integration between spatio-temporal databases and complex event processing.