Ontology-Driven Event Detection and Indexing in Smart Spaces
Li Zang, Chao‐Hsien Chu, Wen Yao, Richard A. Behr · 2010
By riding the tide of sensor technologies, smart space services collects information about its surroundings and internal objects. However, sensor readings are usually at primitive stage and difficult to be viewed and further retrieved since they have little meaning for naive users, who usually prefer to identify activities or state changes using high-level semantics or concepts. To bridge this gap, we propose an ontology-driven event processing framework as part of the middleware for smart spaces. Smart space event ontology (SSEO) is developed to enable semantic indexing, detect machine-processable events and exchange event data between different processes. A model named Smart Space Event Processors (SSEP) maintains and coordinates various event processes (e.g., event patterns, ontology reasoning rules, and machine-learning algorithms) for semantic events in a smart space. An implementation of SSEP-OntoCEP is introduced. The model elaborates on event composition and semantic labeling by combining event composition technology and semantic reasoning into one coherent system. It has been applied in a smart-space infrastructure named SENSIP.