As the number of context-aware applications increases in the real world, it can be quite difficult to deploy such applications in traditional application servers, which are context-agnostics systems. To address this challenge, we propose a novel approach for easing the deployment of contextaware applications into application serversContext is encoded within an OWLdriven knowledge base. We couple this knowledge base with SWRL rules to encode context-awareness thresholds. SWRL rules are not predefined in the application server. They are instead embedded inside the application bundle built by the developer, next to the business logic of the application. At the application deployment time, SWRL rules are extracted to the knowledge base in order to monitor the relevant context for the application to be deployed. At runtime, the context of each session of the application is monitored in the knowledge base. When a rule is triggered (a context-awareness threshold is reached), a broker inside the application server notifies the application so that it adapts its behavior by switching to a more relevant modality. We show how our approach eases the work of developers for building context-aware application by using our context-aware framework.

This paper provides the state of art and hints on how to lay the foundations of an adaptive QoS approach in Next Generation Networks (NGN). The key idea is to provide a model, which would offer one application version or another, depending on the Quality of Service (QoS) negotiated at the session establishment in a NGN. The stake of this research is a better-balanced usage of the network, for maximizing the service offered to the user given his or her network capacities. It encompasses the model for such an implementation in a NGN as IP Multimedia Subsystem (IMS).