Information Technology Reference
In-Depth Information
•
Characterize the limitations and opportunities of movement analysis in
decentralized spatial information systems, and assess the role of CMA in such
systems.
•
Develop the theoretical underpinnings of information processing strategies that
explicitly exploit movement of study objects and/or data capture devices in decen-
tralized spatial information systems.
The chapter is structured as follows. First, Sect.
4.1
introduces the foundations
of wireless sensor and geosensor networks building the basis for the subsequent
discussion of decentralized movement analysis. Then, Sect.
4.2
discusses two settings
where movement in decentralized spatial systems has been studied in the context
of this topic—static sensors monitoring passing mobile objects, and mobile agents
autonomously monitoring their collective movement behavior. Finally, based on this
discussion of two exemplary settings, Sect.
4.3
derives and isolates a set of more
generic principles for decentralized movement analysis.
4.1 Foundations
Advances in distributed sensing and computing technologies offer new, reliable,
and cost-effective means to collect fine-grained spatio-temporal information when
monitoring natural and built environments—so-called wireless sensor networks. The
definition
wireless sensor networks
(WSN) refers to wireless networks of unteth-
ered, battery powered miniaturized computers with the ability to sense, process, and
communicate information in a collaborative way (Zhao and Guibas
2004
). Example
deployments include hazard management (Duckham et al.
2005
), monitoring seismic
activity (Werner-Allen et al.
2006
) or traffic flow (Kellerer et al.
2001
). When specif-
ically monitoring phenomena in geographic space such systems are called geosensor
networks (GSN, Nittel et al.
2004
). Geosensor networks offer a powerful large-scale
alternative to conventional small-scale remote sensing and ground surveys. Figure
4.1
illustrates the basic set-up of a geosensor network.
On top of the many technological challenges keeping the engineers busy, wireless
sensor networks, and in the interest of this volume, geosensor networks also pose
substantial challenges for processing the information generated in such networks.
Whereas conventional geographic information processing is based on centralized
computing models, where sophisticated and powerful databases collate and process
information globally, no such omniscient computing capability exists in geosensor
networks. By contrast, such systems require a new way of spatial computing, where
spatial knowledge is generated from collaborating, but distributed computing nodes
that have only partial knowledge (
P8
. Laube and Duckham
2009
). Much conven-
tional spatial computing nowadays is distributed computing (Worboys and Duckham
2004
; Duckham
2012
). That means that many information systems cooperate syn-
chronously in order to complete a task. For example, for offering location-based
services the mobile devices must communicate with a positioning system and spatial