Performance metrics and routing in vehicular ad hoc networks

The aim of this thesis is to propose a method for enhancing the performance of Vehicular Ad hoc Networks (VANETs). The focus is on a routing protocol where performance metrics are used to inform the routing decisions made. The thesis begins by analysing routing protocols in a random mobility scenario with a wide range of node densities. A Cellular Automata algorithm is subsequently applied in order to create a mobility model of a highway, and wide range of density and transmission range are tested. Performance metrics are introduced to assist the prediction of likely route failure. The Good Link Availability (GLA) and Good Route Availability (GRA) metrics are proposed which can be used for a pre-emptive action that has the potential to give better performance. The implementation framework for this method using the AODV routing protocol is also discussed. The main outcomes of this research can be summarised as identifying and formulating methods for pre-emptive actions using a Cellular Automata with NS-2 to simulate VANETs, and the implementation method within the AODV routing protocol.