Reconfigurable middleware for sensor based applications

Abstract

The pervasive computing paradigm introduced context-aware applications that adapt themselves to their surrounding environment based on context information. Context information may be of different types including sensed context information gathered from a variety of sensors. In pervasive computing user intervention/interactions with their application should be minimised. On the other hand, pervasive computing environments are very dynamic environmentscontext information that supports adaptation decisions may not always be available due to user mobility and potential network and/or sensor failures. As context information supports autonomous adaptation of applications, the provision of context information itself has to be managed by an autonomic system able to both recognise and recover from context delivery failures. This paper presents the architecture of a context management system that is reconfigurable with regard to sensor and/or network failure and can support dynamic discovery and replacement of sensors. One of the components of this middleware architecture, the model and management of sensor descriptions, is described in more detail as it plays an important role in sensor discovery and replacement.

An Autonomic Context Management System for Pervasive Computing

“…In [7], [8], we have already discussed how IEEE 1451 and the SensorML description framework provide a way to describe sensor specifications and data processing. However, a TEDS template describes very limited sensor properties and capabilities (i.e., hardware characteristics, sensor calibration, sensing phenomenon), whereas Sen-sorML can capture detailed information, but the description cannot be processed by resource constrained devices.…”

“…To prove the practicability of our dynamic composition of data processing knowledge on various system configurations, we developed a prototype knowledge base system, which allows us to add data processing models and dynamically compose them together as an interconnected knowledge graph as we described in Section V. We conducted several experimental stresstests (adding, removing of large number of process models and searching of process chains) on various system configurations (from single-board router devices, RB532A 7 , to standard single-core desktop) and knowledge base sizes. Figure 7 shows the experiment results (each averaged from 10 individual tests) in finding a matching process chain for a given context fact type.…”

The Sensor Web: systems of sensor systems

“…The Sensor Web as a typical complex adaptive system is required to exhibit all the characteristics of complex adaptive systems. The internet is also considered a complex adaptive system (Hewitt 1986 Web exists within (Hu et al 2006). If we accept that the internet is an overarching network of interconnected networks and resources, the Sensor Web is a network of sensor networks and resources by the fact that it is part of the internet (Comer 1999 Meta-data plays an essential role in discovering, using and managing both sensors and observation data, as temporal, spatial, and thematic information are required to interpret any observation data correctly.…”