@PhDThesis{ Sig08a,
	author = "Stephan Sigg",
	title = "Development of a novel context prediction algorithm and analysis of context prediction schemes",
	school = "University of Kassel, Chair for Communication Technology, ComTec",
	year = "2008",
	month = Feb,
	abstract = "Context-awareness is an area in which we strive to improve the environment application interaction by the consideration of environmental stimuli and with focus on the application’s reaction. The aim is to extend the knowledge of applications on the requirements of a given situation beyond explicit inputs. This approach is motivated by the general observation that a given environment greatly impacts the intention with which an application is executed. Consider, for example, the ringing proueue?le of a mobile phone. When a user is in a meeting, she will prefer a silent mode while in between two meetings she might be willing to be notiueue?ed of incoming calls by a ring tone. The preference of the ringing proueue?le of the user is situation dependent. In context-awareness, the environmental situation is recorded with the help of sensors. One of the vital parts is then to extract valuable information from this knowledge. This extracted information helps to ueue?ne-tune the reaction to the user needs in an ever more advanced manner. Context-awareness is a broad ueue?eld that can be further partitioned into sub-disciplines that themselves constitute interesting research topics. A context-aware application might be aware of its present, past or future context. Most work is carried out taking present or past context into consideration. A research branch that is by far less intensively studied is the derivation and utilisation of future context. The latter research topic is commonly referred to as context prediction. In this thesis we investigate issues related to the inference of future context and to the utilisation of past and present contexts in this setting. We discuss the context prediction task and identify chances and challenges that derive from it. This discussion leads to a deueue?nition of the context prediction task. The context prediction task might be completed by various context prediction schemes. Dependent on the amount of pre-processing or context abstraction applied to context data, we distinguish between various context prediction schemes. We study context prediction schemes for their prediction accuracy and provide guidelines to application designers as to which prediction scheme to utilise in which situation. Analytical results obtained in these studies are conueue?rmed in simulations in several application domains. For context prediction to be applied widely in various application domains, standardised architectures might foster this development. However, only a few approaches to context prediction architectures have been considered yet. These proposals are static in regard to the context abstraction level. We develop a novel architecture for context prediction approaches that is applicable to arbitrary context abstraction levels. This architecture also allows distribution of components to various interconnected devices in a mobile ubiquitous computing environment. Another aspect of context prediction is provided by algorithms for context prediction. For context prediction algorithms, we review the state of the art and discuss strengths and weaknesses of these algorithms. On the basis of requirements identiueue?ed for algorithms in context prediction scenarios, we are able identify those approaches that are best suited for context prediction tasks. A novel context prediction scheme that was developed in the scope of the thesis is also compared to the existing algorithms. For three exemplary algorithms, we provide simulation results in various application domains.",
	ibrauthors = "sigg",
	ibrgroups = "dus",
	url = "http://www.uni-kassel.de/upress/online/frei/978-3-89958-392-2.volltext.frei.pdf"
}