An event detection service for spatio-temporal applications

Abstract

Sense-and-respond applications form an important class of pervasive-computing applications, serving domains such as driver services, field-force automation, and emergency services. The increasing ubiquity of networked data sources including sensors, mobile phones, online services, databases and data feeds presents novel opportunities for the timely use of the data, which is heterogeneous and available in huge quantities. When these data sources are embedded in the physical world, their location, possibly changing over time, becomes an important part of the context. Business applications will want to respond to the raw data from these sources in diverse ways, in a flexible and scalable manner. We introduce a spatio-temporal event detection service aimed at reducing the costs to application providers by enabling an infrastructure shared by many applications and subscribers. It offers a high-level spatio-temporal programming framework that enables application developers to more easily develop applications based on the sense-and-respond model. © 2006 Springer-Verlag.