Database Reference
In-Depth Information
be enabled via careful tuning (e.g., tuning stop identification and interpretation
to make it efficient even for short stops), and via consistency enforcement for
multiple, correlated annotations and segmentations.
Trajectory Storage
Commercial software is not yet ready to support trajectory data with the new
management facilities that this new type of data requires. At this point trajectory
management is appropriately supported only by research-driven prototypes that
have already reached the level of operational systems. This topic presented the
SECONDO system, which is the most advanced system that has been purposely
built to support mobile data management. Ongoing work in the SECONDO team
aims at extending the model and the system in two major directions. On the one
hand, discussions with ecologists have shown that it is crucial to analyze moving
animals in the context of environmental data such as temperatures, elevation,
and snow extent. These data are available as raster data. Hence it is necessary
to handle raster data together with moving object data in a query. SECONDO's
high-level conceptual model needs to be extended with the data types providing
continuous functions of space and the corresponding operations. A second direc-
tion is parallelization using the MapReduce approach in order to make trajectory
database applications scalable. MapReduce will enable distributed execution of
complex queries by controlling Secondo systems running on many computers in
a network. A different trend is represented by efforts to complement commercial
systems with an external layer providing the functionality required for mobility
data management. The Hermes system is the best representative of this trend.
At what pace the new functionality will be integrated into commercial systems
depends on the DBMS industry.
Similar concerns apply to data warehousing systems, yet the situation regard-
ing trajectory data warehouses is far less advanced than for trajectory DBMS and
research still needs to clearly identify and characterize the extensions needed to
upgrade the data-warehousing paradigm to make it suitable for trajectories, and
eventually implement them efficiently.
Privacy Issues
The privacy solutions that have been discussed in the topic are inherently limited
in scope as they are drawn to target specific privacy goals under well-defined
assumptions about the role of untrustworthy parties and their capabilities. A
challenge for the near future is how to overcome the fragmentation of privacy
technologies to achieve solid conceptual foundations. This question is of vital
importance for future research on privacy. A theoretical framework centered
on the concept of location privacy metric has been recently proposed to deal
with the problem. By quantifying the amount of protection offered by a privacy
enhancing technology, location privacy metrics pave the way to the definition
