OPNET Technologies Institute: Institute of Technology, Sligo, Ireland. www.itsligo.ie
7255 Woodmont Avenue Chief Researcher: Paul Flynn
Bethesda, MD 20814 Research Group: Telecommunication & Network Laboratory (TNL)
Tel: 240-497-3000 Department: Information Systems
Fax: 240-497-3001
E-mail: :info@opnet.comwww.opnet.com/home.html
OPNET is a registered
trademark of OPNET Technologies
© 2001 OPNET Technologies
Background:
The background to the research here at IT Sligo stems from Masters Projects on Decision Support tools in the Telecommunications industry where analysis was carried out on existing wired networks to ascertain their bandwidth requirements etc. This research then progressed to an analysis of future technologies in the wireless telecommunications environment. A number of mobile IP technologies were examined and Cellular IP as proposed by the research team at Columbia University was researched. [1] There are a number of developments that are required for this technology to progress in suitability for mobile use of handheld devices in the near future. One of these requirements is the ability of a local Base Station to have the user profile available prior to his arrival in that area. This is achieved by a movement prediction mechanism that trades processing power against the power required for the handheld device to page the base station for such information.
Aims and Objectives:
The aims and objectives of the research are to come up with a suitable movement prediction algorithm that provides an efficient prediction of user movement in a Cellular IP environment. Many such algorithms are in existence at present for Mobile phone users, most however have a number of limitations. Our research has examined such limitations and the objective is to evolve a prediction mechanism that is specifically suited to the handheld mobile user moving in a multi-cellular environment.
Progress to date:
Using OPNET Modeller for generation of user random movement We have set up a test bed to analyze such movement with a view to examining various existing movement prediction algorithms. Stanford University has also provided the wireless community with movement traces for use in defining movement algorithms. SUMATRA is a trace generator that encompasses several calling and mobility models. [3] Using this movement data we have analyzed a total of over for million actual mobile calls for a twenty four hour trace output. Having also a map of the zones, i.e., what zones were contiguous, geographical location, we analyzed movement based on a number of criteria i.e., location of user, direction of user, random movement etc.
Work still to be done:
Using a combination of actual trace data and random generated movement, We will present some movement algorithms that present successful results as appropriate to the technology involved. These will be arrived at as a result of extensive research of current algorithms and running the test bed to compare and build a comparison with the view to arriving at a new more efficient algorithm for movement prediction in a Cellular IP environment.
References
[1] Valko A, Campbell A, Gomez J, 1999., draft-valko-cellularip-00.txt, Columbia University, May 1999.
[2] OPNET Modeler http://www.opnet.com/services/university/
[3] Stanford University Mobile Activity Traces SUMATRA http://www-db.stanford.edu/sumatra/ (2004)