Information Technology Reference
In-Depth Information
Table 4.1
Mobility mode
Sensor nodes
Static
Mobile
Phenomena
Static
I. Geosensor networks
II. Cluster mining with information
grazing
IV. Decentralized flock detection,
Sect.
4.2.2
What is mobile? The sensor nodes, the studied entities, or both?
Mobile
III. Monitor flows in movement
network, Sect.
4.2.1
•
Lagrangian versus Eulerian perspective
. Is the movement monitored as positions
over time (trajectories) or as times when the object passes fixed checkpoints or
cordons?
•
Mobility mode
. What is mobile? Are static nodes tracking mobile objects, or are
mobile nodes monitoring a static environment, or are mobile nodes monitoring a
dynamic environment?
Whereas the three dimensions set out in Sect.
2.2
only focus on the studied move-
ment process and its embedding in a conceptual space model, in decentralized spa-
tial information systems also the data capture system (i.e. a geosensor network or a
VANET) can be subject to mobility. Since both the study object and the monitoring
system can be on the move, there are four possible mobility modes to be considered
for dynamic decentralized spatial information systems (Table
4.1
). Except the static-
static combination of conventional geosensor networks, all possible combinations
were addressed in this topic, with a focus on modes III and IV. Obviously, the more
dynamic the system, the more difficult become the tasks. Hence, mode IV is expected
to be more challenging than modes II and III (Duckham
2012
).
Following the chapter on “Monitoring Spatial Change over Time” in
Duckham (
2012
), two critical distinctions with respect to the information generated
in a dynamic decentralized spatial information system shall be discussed in more
detail here. The first one refers to the temporal nature of the generated information:
Some systems record histories, others record chronicles (Galton
2004
; Duckham
2012
).
Histories
provide a spatio-temporal record of the states of monitored endurants
(e.g. moving objects) through time.
2
By contrast,
chronicles
provide a record of the
occurrences (perdurants) that happened through time (the occurrence of an object
changing from transit edge
e
i
to
e
j
). This is important because most wireless sensor
systems and hence also dynamic decentralized spatial information systems monitor
histories (snapshots) (Duckham
2012
). Therefore in systems requiring chronicles,
occurrences will need to be inferred from states (Duckham
2012
). It is this seman-
tic enrichment that offers an opportunity for geographically informed algorithms in
decentralized movement analysis discussed in this chapter.
2
Recall the SNAP and SPAN ontologies (Grenon and Smith
2004
).
Endurants
or
continuants
are things that endure through time, e.g. a moving object, this printed book (SNAP ontology).
Perdurants
or
occurrents
by contrast are things that occur in time, e.g. the reader reading this topic
(SPAN ontology).
