Overview
Summary
Vehicular ad-hoc networks (VANETs) are mobile ad-hoc networks whose nodes travel in the road and highway infrastructure already in place. VANETs can be used to spread useful information to drivers and vehicle systems such as reports of stalled traffic or extreme weather conditions. Quick and efficient sharing information would vehicles time to determine an alternate, more appropriate route to reach one’s intended destination.
There are several methods of sharing useful information with motorists that have already been implemented such as dedicated AM radio broadcasts and billboards that advertise road conditions to passing drivers. However, these systems rely on both the limited capabilities of existing infrastructure, such as broadcast range, as well as the action of the motorist to seek out this information on his own accord, by finding and tuning to the relevant radio station. This is hardly an ideal solution, and can even be dangerously distracting to a driver. Another approach is to use existing infrastructure such as a 3G data network, to send and receive reports from vehicles.  This solution would only worsen the already heavy load of traffic upon these networks, especially when considering the great volume of vehicles on the road today. Therefore a widespread data broadcast protocol using existing infrastructure is also an unacceptable solution.
Current algorithms for data dissemination are designed for static or low mobility ad-hoc networks, and become ineffective in a mixed network that contains both mobile and static nodes. The difficulty in sharing this information lies in the high mobility of the vehicular nodes. Because of the varying speeds and unpredictability of movement in a VANET, a connection between two nodes will often be a brief encounter with no guarantee of duration or reliability. Therefore it is extremely important to develop an efficient and reliable algorithm for data dissemination in this type of network.
Our solution is a protocol in which every vehicle would be equipped with a radio transceiver that communicates with static infrastructure transmitters as well as other vehicles on the road. While in the range of an infrastructure node, the vehicle will listen to transmissions from the static node to receive any available up-to-date information. When out of range of any infrastructure nodes, the vehicle will be able to communicate with other vehicular nodes in the network, propagating new information it received from the infrastructure to other nodes in the network, or receiving more recent data from other vehicles. Because of the highly mobile nature of the nodes mentioned earlier, in an effort to increase efficiency our protocol works to estimate the best candidates to communicate with when multiple targets are in range.
In addition to developing a data dissemination algorithm protocol, we are testing its efficiency using both computer simulations and scaled testbed experiments. Currently we are analyzing the effectiveness of the protocol using models for highway, rural, and urban driving environments.