Research Article
Experimental Analysis of Multi-hop Routing in Vehicular Ad-hoc Networks
@INPROCEEDINGS{10.1109/TRIDENTCOM.2009.4976248, author={Jose Santa and Manabu Tsukada and Thierry Ernst and Antonio F. Gomez-Skarmeta}, title={Experimental Analysis of Multi-hop Routing in Vehicular Ad-hoc Networks}, proceedings={2nd International ICST Workshop on Experimental Evaluation and Deployment Experiences on Vehicular networks}, publisher={IEEE}, proceedings_a={WEEDEV}, year={2009}, month={5}, keywords={Ad-hoc Networks Experimental Evaluation Multi-hop communications OLSR VANET Vehicular Communications}, doi={10.1109/TRIDENTCOM.2009.4976248} }- Jose Santa
Manabu Tsukada
Thierry Ernst
Antonio F. Gomez-Skarmeta
Year: 2009
Experimental Analysis of Multi-hop Routing in Vehicular Ad-hoc Networks
WEEDEV
IEEE
DOI: 10.1109/TRIDENTCOM.2009.4976248
Abstract
Evaluation of vehicular ad-hoc networks (VANETs) over real environments is still a remaining issue for most researchers. There are some works dealing with common 802.11 analysis over real vehicular environments, which carry out performance tests to measure the quality of the communication channel and justify results according to physical and MAC conditions. There are only a few works regarding multi-hop experimentation in this field, and even less (if not none) testing multi-hop protocols. In this paper an integral VANET testbed is evaluated, using 802.11b and a multi-hop network managed by the Optimized Link State Routing protocol (OLSR). Up to four vehicles are used over urban and highway environments to study the VANET performance, and different metrics are used to analyse the results in terms of delay, bandwidth, packet loss and distance between nodes. Furthermore, a deeper analysis is carried out to study the route followed by packets end to end, which enables us to count the number of hops and detect the links where packets are lost. Because a routing protocol is used, results differ from traditional two-hop and staticroute tests, presenting a more realistic study. OLSR is considered as a good reference point for the research community, although it is not the most suitable protocol for vehicular environments, as results show.
