Network-based Generator of Moving Objects


The evaluation of spatiotemporal database systems or their components requires the definition of suitable benchmarks simulating the typical behavior of moving objects. Previous approaches do not consider one important property of moving objects: In many applications dealing with spatiotemporal data, the objects follow a given network. Therefore, benchmarks require datasets with such "network-based" moving objects. Important aspects for the generation of such objects are the maximum speed and the maximum capacity of connections, the influence of the other moving objects to the speed and the route of an object, the adequate determination of the start and destination of an object, the influence of external objects and events, and time-scheduled traffic. This observations led to the development of this generator for spatiotemporal data. It combines real data (the network) with user-defined specifications of the properties of the resulting dataset.


Examples and Resources

Current Version: 2.1.2, 2005-04-19

Because of the bug fixes and improvements, it is strongly recommended not to use the versions 1.x and 2.0.


Note for producing large sets of moving objects:

  • Disable the visualisation in the property file (and don't set a waiting period)!
  • If necessary, increase the memory allocation pool of the Java application launcher.

Data Files (for all versions)


Frequently Asked Questions

What do the output columns mean?

The reporting depends on the Reporter class that is used and the type of the output file. The DefaultDataGeneratoruses the class PositionReporter (the source code of both classes and of other Reporter classes are included in the source code download). The type of output file is determined by the contructor of the Reporter class considering the file extension. The two print methods determine the reported parameters. For details, take a look into the source code of the relevant R