Ontology-based context information modeling for smart space

The contexts in highly dynamic environments are rich and changeable. Thus context representation, usage and management become very important. The paper utilizes a kind of context model based on ontology for smart space. The context model is composed of the top layer ontology model and the specific field ontology model. The top layer ontology model is the abstract of context information in smart space; on the other hand, the specific field ontology model is the embodiment of the top layer ontology model in the specific field. The top layer ontology consists of classes and attributes. As far as classes, we have introduced seven kernel classes to describe the most of contexts in smart space, which are Person, Location, Time, Activity, Physical Entity, Virtual Entity, and Service. The classes in specific domains are the extensions and subclasses of the seven kernel classes. The attributes of classes can be divided into two categories: static attributes and dynamic attributes. We have utilized three meta- attributes which are accuracy, probability and freshness to describe and qualify dynamic attributes.

1. Introduction

Recently, pervasive computing has entered people’ life gradually and played an important role in the daily life. The openness, mobility and changeability of pervasive computing are increasingly obvious. Smart space is a representation of pervasive computing applications. However, the contexts in s mart space are va rious and changeable. The context modeling, management, representation and reasoning become the indispensable basis of the smart space's research.