Environment Model

As described in [1], PeerfactSim.KOM requires information about obstacles, pathways, and waypoints to model the environment for a simulation. This information can either be obtained by reading SVG-images or parsing the data from OSM. Based on the two procedures, purpose-based and realistic environments can be created as described in the following.

Image-based Modeling of the Environment

PeerfactSim.KOM relies on common SVG-images to create purpose-based environments. The simulator uses the XML-based data representation of SVG-images to extract obstacles, pathways, and waypoints. The prefix of an ID, which is assigned to each element, specifies the type of the element in the modeled environment. Valid prefixes of an ID comprise map, way, and obstacle.
The way-elements are used to specify the pathways through the map. The endpoints of a pathway as well as the junctures of the connected SVG-paths are used to identify the waypoints, which serve as destination as well as for path calculation.
The obstacles are defined by elements, whose prefix of the corresponding ID matches the term obstacle. The corresponding element either specifies a rectangular obstacle or consists of several SVG-paths, which are used to create an obstacle of a polygonal shape.

OSM-based Modeling of the Environment

OpenStreetMap provides the corresponding data in an XML-file, which consists of so-called OSM-nodes, -ways, and -tags. Out of this information, PeerfactSim.KOM generates and displays the identified elements. OSM-ways are used to connect different OSM-nodes, which belong together. The corresponding OSM-tag classifies the resulting object, such as obstacles, paths, or waypoints.
OSM-nodes and -ways that are marked with the building-tag, are used to create obstacles. An additional OSM-tag, denoted as amenity, specifies the type of
the building to determine the popularity of that building. PeerfactSim.KOM uses this
popularity to define attraction points and influence the movement of nodes.
For the creation of pathways, PeerfactSim.KOM parses OSM-ways, which are marked with the highway-tag. The identified OSM-ways are used to create a pathway, while the adjacent OSM-nodes represent weak waypoints. Besides the weak waypoints, there exist also strong waypoints. Strong waypoints specify the next destination of a node, whereas weak waypoints are only used to calculate the path to the selected destination. PeerfactSim.KOM provides the OSM Hotspot Strategy, the Weak Waypoint Strategy, and the SLAW Waypoint Model [2], which is adapted from the implementation from BonnMotion to place strong waypoints on a map.

Examples of a SVG-based map [1].
Examples of an OSM-based map [1].

Mobility Model

The environment serves as input for the mobility model. In PeerfactSim.KOM, a mobility model consists of two strategies: a waypoint selection strategy and a local movement strategy. The waypoint selection strategy is used to select the next destination of a node, while the local movement strategy determines the path from a node’s current position to its destination. Depending on the modeled environment and the available information on the map (e.g. obstacles, hotspots, or waypoints), appropriate models for both strategies can be selected.

The simulation platform currently implements three way-point selection strategies that select a node’s next destination. These strategies are derived from existing movement models covering (i) the SLAW Waypoint Model [2], (ii) the Random Waypoint Model [3], and (iii) the Gauss-Markov Mobility Model [4].

The local movement strategy is applied to guide a node from its current to the selected location. PeerfactSim.KOM provides two strategies. The Linear Movement strategy provides a basic mobility model, where nodes move on a direct line between waypoints. The Shortest Path Waypoint movement requires the provided weak waypoints to determine the shortest path from a node’s current to its selected position.

References

[1] D. Stingl, B. Richerzhagen, F. Zöllner, C. Gross, R. Steinmetz. PeerfactSim.KOM: Take it Back to the Streets (accepted for publications). HPCS. 2013

[2] Lee, K., Hong, S., Kim, S. J., Rhee, I., & Chong, S. SLAW: A New Mobility Model for Human Walks. IEEE INFOCOM (pp. 855–863). 2009

[3] D. B. Johnson and D. A. Maltz, “Dynamic Source Routing in Ad Hoc Wireless Networks,” in Mobile Computing, 1996, vol. 353, pp. 153–181.

[4] B. Liang and Z. J. Haas, “Predictive Distance-Based Mobility Management for PCS Networks,” in IEEE INFOCOM, 1999, pp. 1377–1384.

A A A | Drucken | Impressum | Sitemap | Suche | Kontakt
Zum SeitenanfangZum Seitenanfang