Directed Waypoint Model: A Hybrid Approach to Realistic Mobility Modeling in Mobile Ad-hoc Networks
No Thumbnail Available
Files
Date
2005-08-08
Authors
Journal Title
Series/Report No.
Journal ISSN
Volume Title
Publisher
Abstract
Researchers are highly dependent on simulations to compare and evaluate the performance of Mobile Ad-hoc networking protocols. Mobility modeling plays a very important role in simulation-based studies of MANET protocols, since the mobility of nodes is the limiting factor in the performance of such protocols. Currently, simulations of MANET protocols rely on simplistic, purely stochastic models such as the random waypoint. These models do not reflect the motion of real world MANET nodes, which may be deployed in diverse scenarios. At the present time, the only available alternative to these models is fine-grained detailed simulations based on extensive surveys of actual node behavior, or extensive sets of real world traces. This work proposes an alternative hybrid mobility model that extracts scenario-specific information from real traces and generates realistic node movements for particular scenarios. The distinguishing features of our model are its ability to reproduce spatial-temporal node movement properties from sample traces and its general adaptability to traces from any real-life scenario. We develop a framework for our model and apply it to reference traces from several realistic scenarios. The hybrid model is then evaluated for accuracy by comparing the performance of MANET routing protocols using random waypoint with realistic parameters and our model against the original reference traces used by our model. The results show the relevance of accounting for spatial and temporal aspects of node movements in mobility models.
Description
Keywords
mobility modeling, MANETs, MANET routing protocols, detailed simulation frameworks, stochastic mobility models, simulation modeling, spatial and temporal dependencies, performance evaluation
Citation
Degree
MS
Discipline
Computer Networking
