Implementation and performance assessment of location-based routing protocols for MANETs
Document typeConference lecture
PublisherAssociation for Computing Machinery (ACM)
Rights accessRestricted access - publisher's policy
Mobile ad-hoc networking is a topic that has caught the interest of the research community for years. Although MANETs were introduced time ago, routing in such difficult environments is still an open issue. Routing protocols have been presented to deal with dynamic route building, including proactive and reactive approaches. The newest proposals are often based on positioning, owing to the clear benefits that this approach offers to overcome the excessive flooding overhead that routing protocols usually introduce. This paper compares the performance of three routing protocols: DYMO, DYMOselfwd and AODV-Line. The first is included for historical reasons, since it was the reference protocol for MANET routing for many years. DYMOselfwd is a novel routing protocols that trades-off the simplicity of DYMO and the scalability of location-based routing protocols. Finally, AODV-Line is chosen as an example of greedy location-based routing protocol. These three protocols have been implemented in OMNET++, a general-purpose event-driven simulation tool freely available for the research community. This paper provides a thorough assessment of the performance achieved by these three protocols under different conditions depending on: 1) the node density (i.e. sparse and dense networks), 2) the mobility pattern (i.e. pedestrian and vehicular nodes) and 3) the payload length (e.g. data coming from sensors wrt. regular data traffic).
CitationMartin-Escalona, I., Zola, E., Perrone, F., Barcelo, F., Machado, S. Implementation and performance assessment of location-based routing protocols for MANETs. A: ACM international symposium on design and analysis of intelligent vehicular networks and applications. "DIVANet '17: proceedings of the 7th ACM Symposium on Development and Analysis of Intelligent Vehicular Networks and Applications". New York: Association for Computing Machinery (ACM), 2017, p. 125-134.