Distributed Context Service Architecture for Heterogeneous Radio Environments

Abstract:
State-of-the-art mobile phones ("smartphones") have become all-in-one devices whose capabilities range from well-known voice services to multimedia applications and navigation systems. Moreover, they host a broad variety of sensors for contextual data (light, noise, acceleration, orientation, GPS) as well as software performing complex operations on these context data, e.g. situation recognition. However, there is no established framework (neither from technical nor from a business perspective) for providing other, less advanced terminals a remote access to these context functionalities residing on peer devices. Designing a scalable distributed context service registry for providing context data and value-adding context services is the challenge addressed in this paper. An according implementation supporting mobile as well as fixed nodes will be presented. The unique feature of the implementation is that it will overcome the weaknesses of current (centralized) context management platforms by applying the logic of distributed hash tables (DHT) to create a highly available and reliable decentralized context service registry.