NeTS-NBD: Measurement-Based Mobility Modeling for MANETs

North Carolina State University, Raleigh NC

Investigators

Abstract

NeTS-NBD: Measurement-Based Mobility Modeling for MANETs Award 0626850 Mihail Sichitiu Mobile ad hoc networks (MANETs) are inherently cooperative. Because possibly all communicating devices may undergo varying degrees of mobility in MANETs, they rely heavily on nearby nodes to maintain network connectivity. Therefore, the underlying mobility patterns of mobile nodes strongly influence the performance of MANET protocols. Modeling and reproducing the mobility patterns arising from real scenarios are very important for simulation-based performance studies of MANET protocols. The primary goal of the study is mobility modeling and generation. In contrast to existing work, this research focuses on modeling the mobility patterns from real scenarios and emulating such patterns in synthetic mobility traces that can be used for creating realistic simulation environments. The proposed model has the desirable characteristics that it is customizable to match any real scenario (e.g., busses in a city, students in a campus, or zebras in a herd), while allowing for convenient diversification (i.e., allows changing the number of nodes, density, etc.). The main deliverables of this project are a suite of mobility models and the tool that uses them to generate the traces. Since for the foreseeable future MANET performance evaluation will be based on network simulations, it is expected that the results of this project will be widely used in the MANET community. It is envisioned that the proposed model will effectively replace the random way point as the standard mobility model used in any MANET performance evaluation.