You are here

Vehicular Network

Vehicular Netowrks are a kind of MANETs in which nodes are vehicles that follow particular mobility patterns regulated by vial normatives. In the last years, several organizations have supported standardization activities such as IEEE 802.11p and WAVE (Wireless Access Vehicular Environment), ISO TC204 WG-16 CALM architecture, ETSI-ERM TG37, applied to vehicular communications. Organizations such as C2CCC (Car to Car Communications Consortium) have brought the European vehicle manufactures to join in an initiative with the objective of further increasing road traffic safety and efficiency by means of inter-vehicle communications.

Communications in VANETs mainly may be of two types: single-hop to neighbor cars to advise of an event (e.g. braking) or multi-hop to either disseminate information or to query for a service. From the application perspective, applications can be categorized as Transportation-related applications and Convenience and personalized applications.

As in MANETs, communications in VANETs may be impacted by several factors:

  • High velocity of the vehicles
  • Environment factors: obstacles (e.g. urban scenarios), tunnels, traffic jams, etc.
  • Determined mobility patters that depends on source to destination path and on traffic conditions
  • Low duration of the communications (aprox. 4-6 hops)
  • Intermittent communications (isolated networks of cars due to the fragmentation of the network)
  • High congestion channels

These factors result in a network with frequent fragmentation in small parts, rapid changes in the network topology due to high node speeds, small effective network diameters and limited redundancy. In fact some papers show that at nine hops, before a request packet may be ACK, the path has dissapeared. This makes that proactive and reactive MANET protocols behave poorly in a VANET scenario.

The Research group has been participating in the following research topics related to VANET's:

  • Dissemination of information in VANET's. Several of our works consider the dissemination of information in a vehicular network with disruption. In such scenarios, nodes can not find forwarders and have to store-carry-and-forward data in a Disruptive Tolerant Network environment. In this case, the use of Epidemic Modeling is quite useful. Such models consider that nodes exchange data whenever they have the opportunity to meet another node. In the mean time, they carry the traffic. Under these assumptions, we have work in the cooperative download of traffic in urban, sub-urban and highway scenarios.
  • In the VNET-2 and VNET-3 projects, we studied routing algorithms applied to safety communications and how to bring applications to Delay Tolerant environments. One of the outcomes of the project was a cooperative mechanism (C-ARQ) in which vehicles that enter under the coverage of an AP, were caching data addressed to other vehicles. When the vehicles leaved the coverage area, each node requested the missed data to other vehicles. We have developed a prototype including C-ARQ (Cooperative-ARQ) and tested how losses may be recovered in vehicular scenarios.
  • We also worked, in the VNET-3 project, in finding those locations in which RSU's (Road Side Units) could be placed in an urban scenario. The idea was that using Linear Programming (LP), we could define those locations that aimed to deliver data to the maximum number of vehicles using the minimum number of RSU's.
  • In the e-VENET (Electrical Vehicle Networks - Future Networking for Low Emission Human Mobility, Nov 2011- May 2012) project, we addressed the need to support the deployment of charging stations with a communication infrastructure and communication protocols that can timely advertise location, availability and waiting time of charging stations, and that were capable of interacting with on-board navigation systems.