Information stream oriented content adaptation for pervasive computing

Content adaptation is a challenging work due to the dynamism and heterogeneity of the pervasive computing environments. Some researchers address this issue by organizing services into customized applications dynamically. However, due to the maintenance of the dependencies between services, these systems become more complicated with the growing of the system scale. Programming for these systems is also error-prone. This paper introduces our work in this field, UbiCon, an information streams oriented content adaptation system. By abstracting information streams into generic CONTENT entities, the system provides a simple and powerful means for services to operate information stream. The CONTENT is created by the system dynamically, and essentially has local association with related services. As a result, the CONTENT is also used as a loosely coupling mechanism for cooperating associated services. By abstracting services with a T model, the services effectively cooperate together with other services. As a result, a collection of sophisticated applications can be built with this services model. As a proof of concept, we have developed a prototype implementation. The preliminary experiments show the effectiveness of this system.