A Semantic Context Management Framework on Mobile Device

We present a semantic context management framework named ContextTorrent, which can make various types of context information be semantically searchable and sharable among local and remote context-aware applica-tions. We implement this framework on the Google An-droid platform with its elegant application support. An open source RDF parser has been extended to effectively get RDF triples from files or over the network. Three em-bedded database systems were evaluated for storing ontol-ogy represented contexts in the resource-constrained mobile devices. We use the FOAF ontology schema and a synthetic data set of up to 2500 records to evaluate the context query and storage performance. Ordinary context queries can be replied instantaneously.