Pervasive Computing environments are dynamic and heterogeneous. They are required to be self-managing and autonomic, demanding minimal user’s guidance. In pervasive computing, contextaware adaptation is a key concept to meet the varying requirements of different clients. In order to enable context-aware adaptation, context information must be gathered and eventually presented to the application performing the adaptation. It is clear that some form of context categorization will be required given the wide range of heterogeneous context information. Categorizations can be made from different viewpoints such as conceptual viewpoint, measurement viewpoint, temporal characteristics viewpoint and so on. To facilitate the programming of context-aware applications, modelling of contextual information is highly necessary. Most of the existing models fail both to represent dependency relations between the diverse context information, and to utilize these dependency relations. A number of them support narrow classes of context and applied to limited types of application, and most do not consider the issue of Quality of Contextual Information (QoCI). Along with a detailed context categorization, this paper will analyse existing context models and discuss their handling of dependency issues. It uses this analysis to derive a methodology for quality context information modelling in context aware computing.
|Title of host publication||Proceedings of the IJCAI 2005 workshop on AI and Autonomic Communications|
|Number of pages||10|
|Publication status||Published - 2005|