Towards contextaware and resource-driven self-adaptation for mobile handheld applications

Mobile handheld computing is gaining momentum as more and more wireless handheld devices are being used to ac-complish various computing tasks. Context-awareness can help to adapt and personalize applications or to optimize resource usage. However, many existing context infrastruc-tures by themselves impose heavy resource requirements for deployment or highly affect the limited autonomy of the mo-bile device. In this paper we describe how resource-driven self-adaptation is used in our layered context-driven applica-tion middleware in order to enable deployment on a mobile device and optimize resource usage. Our evaluation shows that the introspection and intercession capabilities of the self-adapting middleware provide the necessary flexibility to achieve a balanced resource usage between the context-driven application middleware and the context-aware appli-cations, and that the resource-driven self-adaptation is able to more than double the battery lifetime in real life scenar-ios