Lightweight Ontology-Driven Representations in Pervasive Computing

Abstract

A clearly specified representation of diverse entities is needed to refer to them in pervasive computing applications. Examples of such entities include physical objects, operations, sensor and actuator resources, or logical locations. We propose a novel way to systematically generate representations of entities for programmable pervasive computing platforms made of tiny embedded nodes. Our original idea is to generate a very lightweight, though semantically-rich, representation from a possibly complex ontological specification. At the platform development phase, a domain ontology is used to describe the target environment. A preprocessing tool produces the ontologydriven, lightweight representation, which comes in two flavors: a humanreadable one, to be used for programming, and a binary one, to be used at runtime. Our approach makes it possible to take advantage of all the benefits of ontology-based modeling and, at the same time, to obtain a representation light enough to be embedded in even the tiniest nodes.

A comprehensive engineering framework for guaranteeing component compatibility

“…The ODA class is a complex structure that is constructed by combining basic concepts from a domain ontology. It extends our previous work presented in Domaszewicz and Rój (2005).…”

ROVERS: Pervasive Computing Platform for Heterogeneous Sensor-Actuator Networks

“…An ontology preprocessor tool is used to generate a full (possibly huge) collection of instructions and events for the domain (samples are given in Table 1). Our initial work on this novel approach has been presented in [3]. …”

“…Relatively little work has been done in this area. An original idea included in the programming model and requiring further work is the derivation of lightweight programming artifacts (events and instructions) from an ontology [3]. More work is also needed on how to better support writing pervasive computing applications for an unknown mix of nodes.…”