Information Technology Reference
In-Depth Information
ment is easily available, the variety of applications
that can benefit is only limited by the imagination.
The information can be collected by sensors which
are embedded in devices and systems hidden in
the environment, and the smart space technol-
ogy may support their interoperability, as well as
the abstraction of semantically rich information
from the data collected from the environment.
Emerging mobile devices already benefit through
the addition of sensors to their set of traditional
conventional resources. Sensors, such as cameras,
compasses, gyroscopes, accelerometers, GPS,
RF based devices and many others may provide
inputs to the smart spaces. The smart spaces only
take care of information interoperability, whilst
connectivity, smart space and service discovery
are orthogonal issues. Furthermore, as shown in
Figure 1 (bottom), smart spaces envisage a clear
separation between the data and applications. A
solution introduced in (Lassila 2007) implements
this concept with ontologies using the Resource
Description Framework (RDF) mapped onto a
common data model represented as a graph. An
ontology is a shared knowledge standard or a
knowledge model defining primitive concepts,
relations, rules and their instances, which comprise
topic knowledge (Zhou 2005). Ontology can be
used for capturing, structuring, and enlarging
explicit and tacit topic knowledge across people,
organizations, and computer and software systems
(Edgington et al. 2004). In smart spaces, ontolo-
gies can be used to describe the semantics of a
space and the semantics of applications, services,
data and the context where they are used. The
standard set of Semantic Web languages (mainly,
RDF and OWL) provided by the World Wide Web
Consortium, represents the most widely adopted
solution to implement ontologies.
A platform which is hosting a smart space may
be very simple, as it may just require a service
consisting of a repository for storing and manag-
ing graphs and a path query language for graphs
with reasoning capabilities, where the reasoning
role is to extract (deduct) information from the
graph, which is not explicitly stated (Lassila
2008). This service is the core component of the
interoperability platform considered in this chapter
for our discussion on smart space design and it
goes a long way in the direction of making smart
space programming an easy task.
The Interoperability of Smart Spaces
From a technical viewpoint, interoperability is a
property of computational units that makes them
able to inter-operate. For smart spaces that rely
on legacy systems and devices and their ability to
work together to achieve a common goal, interop-
erability is a prerequisite that has to be fulfilled.
That is only possible if the interacting units use
the same interaction model at every abstraction
level. Due to its objective, this interaction model
is called an interoperability model.
The interoperability models proposed in lit-
erature are diverse; the levels of abstraction are
different, and they differ in the methods applied to
and the technical solutions used for achieving the
interoperability. When comparing the maturity of
five interoperability models, the following most
significant potential, concerns and barriers were
identified (Guédria et al. 2008):
•
The use of standards creates potential
(openness) and is addressed in every in-
teroperability model. Thus, the use of stan-
dards provides advantages for open smart
spaces.
•
Data and service interoperability are the
concerns of smart spaces. Data interop-
erability is addressed in LISI (Levels
of Information System Interoperability)
(C4ISR Interoperability Working Group
1998) and LCIM (Levels of Conceptual
Interoperability Model) (Tolk & Muguira
2003). Only LISI concerns service
interoperability.
•
Conceptual and technological barriers
were identified in two (LISI and LCIM)
Search WWH ::
Custom Search