Routing in highly dynamic wireless networks such as Vehicular Ad-hoc Networks (VANETs) is a challenging task due to frequent topology changes. Sustaining a transmission path between peers in such network environment is difficult. In this thesis, Previous Hop Routing (PHR) is poposed; an opportunistic forwarding protocol exploiting previous hop information and distance to destination to make the forwarding decision on a packet-by-packet basis. It is intended for use in highly dynamic network where the life time of a hop-by-hop path between source and destination nodes is short. Exploiting the broadcast nature of wireless communication avoids the need to copy packets, and enables redundant paths to be formed. To save network resources, especially under high network loads, PHR employs probabilistic forwarding. The forwarding probability is calculated based on the perceived network load as measured by the arrival rate at the network interface. We evaluate PHR in an urban VANET environment using NS2 (for network traffic) and SUMO (for vehicular movement) simulators, with scenarios configured to re ect real-world conditions. The simulation scenarios are configured to use two velocity profiles i.e. Low and high velocity. The results show that the PHR networks able to achieve best performance as measured by Packet Delivery Ratio (PDR) and Drop Burst Length (DBL) compared to conventional routing protocols in high velocity scenarios.