Geography Reference
In-Depth Information
Fig. 6.1
Sensors periodically report time-variant traffic volumes on Minneapolis-St Paul high-
ways (Best viewed in color, Source: Mn/DOT) (Color figure online)
Tabl e 6. 1
Example queries in time-varying networks
Category
Example questions
Snapshot
Where are the evacuee groups located at present?
Node/edge centric
Which evacuee groups are at a node
Time-aggregated
At different time instants?
Trajectory based
Find the route of a person on a given day
6.2.3
Knowledge Discovery in Spatio-temporal Sensor
Networks
Many scientific domains collect sensor data in outdoor environments with underly-
ing physical interactions. A collection of sensors may be represented as a sensor
graph where the nodes represent the sensors and the edges represent selected
relationships.
For example, sensors upstream and downstream in a river may have physical
interactions via water flow and related phenomena such as plume propagation.
Relationships can also be geographical in nature, such as proximity between the
sensor units. A sensor graph is spatio-temporal in nature since the relative locations
of the sensor nodes and the time-dependence of their characteristics are significant.
Developing a model that facilitates the representation and knowledge discovery on
sensor data is extremely critical in mining useful information from data.
Search WWH ::
Custom Search