Realistic Mobility Modelling for Mobile Ad Hoc Networks

Overview |  People |  Papers |  Software |  Back to Research 

Overview

Simulation is a valuable research tool and the field of mobile ad hoc networks uses simulations extensively to evaluate designs. However, better mobility models are required since the research community has been using the unrealistic random waypoint model for very long. With the emergence of Vehicular Ad Hoc Networks the research community requires a mobility model which can capture at least some of the features of vehicle mobility. The objective of this project is to provide better mobility modelling for practical research in Mobile Ad Hoc Networking.

We provide a realistic mobility model based on vehicular traffic. We make use of the publicly available TIGER (Topologically Integrated Geographic Encoding and Referencing) database from the U.S. Census Bureau, giving detailed street maps for the entire United States of America, and model automobile traffic on these maps. Our tool generates scenario files which are compatible with the scenario file format of the ns-2 simulator.

People

• Faculty: David B. Johnson
• Graduate students: Amit Saha

Papers

• Amit Kumar Saha and David B. Johnson. Modeling Mobility for Vehicular Ad Hoc Networks. Appeared as a poster in the First ACM Workshop on Vehicular Ad Hoc Networks (VANET 2004), Philadelphia, Pennsylvania, Oct 2004. Here is the full paper .

Software

• The TIGER database files for 2003 as well as required documentation are available online from the 2003 TIGER/Line File Download Page. It seems that the 2003 data is no longer available for download and have been replaced by 2004 data. There might have been some format changes in the newer data files and so my scripts might not run directly on these. However, you should be able to modify a single script, "chop.bash" in order to get the script to work with the new files. From a first glance of the new data format, it seems that the relevant fields and their positioning and format inside the data file has not changed from the perspective of the script, i.e. "chop.bash" should run unmodified on the newer data files as well, but there might be some modifications needed.
• Tar ball of the software. Use 'tar -xvzf Vanet2004.tgz' to untar and gunzip. If that does not work then use 'gunzip Vanet2004.tgz' followed by 'tar xvf Vanet2004.tar'. After extracting the tar ball there should be three directories, Vanet2004/, Vanet2004/Scripts/, and Vanet2004/Scripts/Examples/.
• If you want just the scenarios as mentioned in the Vanet2004 paper then here they are:
  • Data file corresponding to West University area, Houston, TX, USA.
  • Postscript file corresponding to West University area, Houston, TX, USA.
  • Data file corresponding to Afton Oaks area, Houston, TX, USA.
  • Postscript file corresponding to Afton Oaks area, Houston, TX, USA.