| |
5
General Concepts, Toolkits,
and Frameworks
In order to apply Xtreme Prototype in practice re-usability is a major prerequisite.
This chapter discusses the state-of-the art (see Section 5.1) in ubiquitous computing
research and proposes two concepts that should guide and streamline work processes
during the application of Xtreme Prototyping.
Section 5.2 develops a conceptual framework, referred to as the sensor-opportunities-
based approach, that allows to categorize existing sensors and evaluates their utility
in various applications. Eventually, this framework shall guide application design-
ers to choose meaningful sensor subsets, inspire new systems, and evaluate existing
applications. Furthermore, this section describes a conceptual categorization frame-
work of sensors and reviews existing ubicomp applications using the framework.
Finally, an evaluation of sensing technology with respect to the framework is pre-
sented.
Section 5.3 proposes a model for perception as part of a layered architecture. When
looking at current implementations and when interviewing application developers
it becomes evident that sensing and distributed perception is re-invented and re-
implemented over and over again in various systems. Often the argument is that
the application has very specific requirements (e.g. power consumption, response
time) and that therefore a particular, often monolithic architecture and specific
implementation is the best solution. While understandable from an individuals
point of view this is clearly an undesirable status quo. In particular, as we move
from mere demonstrators within controlled research labs to larger scale applications
in more open environments there is a stringent need for architectural frameworks
including layers of abstraction, modularity, and reusability. While such an overall
architectural framework obviously is a long-term research goal, this section proposes
a model for perception as part of such an architecture.
61
|  |
|
| |
|
|