Authors: Filip Perich

Date: August 12, 2002

Abstract: We present an architectural support for building context-aware applications. With the increasing acceptance of dynamic mobile and ubiquitous computing environments, applications are no longer limited to relying only on the information explicitly provided by users. Instead, these environments invite a new kind of applications leveraging from implicit information that can range from a simple time-of-a-day information to information about user location to even an elaborate description of user current activity. This implicit information is known as the context, and although the context-aware computing paradigm has been introduced over a decade ago, so far there exist only a limited number of context-aware applications available to everyday users. We believe that this is due to the tight coupling of contextual operations where each application is required to sense, collect, interpret and apply all information independently from others. To overcome this issue we propose a decoupled approach, where the information sensing, aggregation, and interpretation is provided by external services, and applications are concerned with only applying the contextual information to their needs. Moreover, we focus on the design and specification of a generic service capable of both aggregating and interpreting context information as part of the Me-Centric computing project, and demonstrate its capability in an office-based scenario.

Type: TechReport

Address: Palo Alto, CA

Institution: Hewlett-Packard Laboratories

Google Scholar: search

Bookmark at: Digg | Del.icio.us | Connotea | CiteULike