SUP-NHD-Strategiske UoH-progr.fin.NHD

SIRIUS focuses on three areas in which it aims to come up with new research results to considerably forward the state-of-the-art within pervasive systems development: 1. Sensing, through a set of "high-level" services that shield system developers and ap plication programmers from the complexity and heterogeneity of collecting and processing sensor data. In order to simplify accessing and processing, we propose to develop middleware-like sensing services that have declarative interfaces and provide transp arency. One of the core challenges is to make these services extensible such that multiple application-specific data models can be used, possibly for the same data source. 2. Adapting, in form of an adaptation framework that enables autonomic adaptation of individual system components in large-scale pervasive systems. Adaptation of a single component will often impact other system components. This could in the worst case lead to oscillating system behavior or cascading effects. The core challenge is to f ind efficient adaptation strategies that even with partial knowledge about the system can minimize the negative effects of self-adaptation on other components. 3. Protecting, by tools that allow the system to detect adverse conditions and anomalies in t he system. Once an adverse condition is detected it is necessary to perform a diagnosis to find the root cause problem, derive countermeasures and apply them later on as an adaptation to a known adverse condition. The core challenge is to minimize the hum an assistance in this process with the ultimate goal of self-learning components that generate new knowledge. Sensing, adapting, and protecting will be applied within and for an information space for rescue and emergency support. This is a massive-scale information and communication platform that provides means to share arbitrary kinds of information among distributed services and applications in a dynamically changing mobile ad-hoc pervasive environment.