Context reasoning using extended evidence theory in pervasive computing environments
Future Generation Computer Systems
Review: Situation identification techniques in pervasive computing: A review
Pervasive and Mobile Computing
Middleware for pervasive computing: A survey
Pervasive and Mobile Computing
Decentralized checking of context inconsistency in pervasive computing environments
The Journal of Supercomputing
| Hi-index | 0.00 |
Context, the pieces of information that capture the characteristics of ubiquitous computing environment, is often imprecise and incomplete due to user mobility, unreliable wireless connectivity and resource constraints. While many context reasoning schemes have been proposed to assist ubiquitous applications, these schemes often suffer from the assumption that the contexts are complete and precise. The main challenge for context reasoning is how to interpret and infer contexts from existing contexts that are imprecise and incomplete. To this end, we propose the DSCR脳 approach — extended Dempster-Shafer theory for Context Reasoning, which applies Dempster-Shafer theory to ubiquitous computing environments by following a new context-aware architecture. DSCR脳 solves the fundamental problem in Dempster-Shafer theory – intensive computation through evidence selection strategy. This strategy takes advantage of the k−l algorithm to select evidence with the highest beliefs, which considerably reduces the computation overhead. The proposed approach is evaluated through extensive experiments. The results show that DSCR脳 is appropriate to reason contexts from incomplete and imprecise contexts in ubiquitous computing environments.