GeoTAR:A Map-based And Traffic-aware VANET Routing Protocol

With the high demand for vehicular communication,VANET(Vehicle Ad hoc Networks)has become a promising technology to enable ITS(Intelligent Transportation Systems)by providing efficient V2V(Vehicle to Vehicle Communications)and V2I(Vehicle to Infrastructure Communications).Yet,due to the high mobility and unreliable channel quality,VANET has its unique characteristics which pose many challenging routing issues,such as traffic awareness,channel quality assurance,infrastructure utilization.In this thesis,we propose GeoTAR,a map-based and traffic-aware routing protocol,and the main contributions are as follows:(1)In GeoTAR,we utilize the digital map to assist routing decisions.Different from the previous research work,we divide the digital map into grids by Geohash,and set multiple precision levels to collect road information and forward data packets.Moreover,the location information of a node is no longer represented by twodimensional pair of latitude and longitude,but the one-dimensional Geohash code,which not only reduces the transmission overhead of the beacon packet,but also facilitates the matching of the node and the grid,and also simplifies the internode distance calculation.(2)Through Geohash,the roads are meshed and each node is matched with the road grid.We can collect the distribution of nodes on the adjacent road segments periodically through dedicated collection package,and calculate the road conditions of the corresponding road segments statistically at the intersections.When there are data packets to be forwarded,the local traffic information dynamically collected is used to decide routing path.In order to improve the network throughput by using the channel resources of the roadside units,we reflect the distribution of the roadside units into the digital map,and set a virtual road segment between the intersections where the roadside units are distributed.We manually set the road conditions of the virtual road segments so that the channel resources between the roadside units are prioritized when performing routing path calculation.(3)In terms of link quality control,the unidirectional link problem is common due to the influence of buildings on the wireless transmission at the intersection.To avoid the impacts of the buildings,we select the forwarding node at the intersection through the Geohash grid when forwarding near the intersection.In addition,we control the single-hop forwarding distance through the grid to ensure the link communication quality when forwarding on the road segment.In NS-3,We first designed a validate experiment to show that selecting node at the intersections preferentially can alleviate the impact of buildings and improve channel quality,and traffic collection can alleviate link breaks on routing paths.Then we implemented a larger-scale experiment.By comparing with conventional routing protocols,we explained that utilizing intersection-first restricted forwarding strategy and high-quality channel resources of the roadside units,GeoTAR has a more stable link quality,higher delivery rates and lower latency.Moreover,with dynamic traffic collection,GeoTAR can maintain high packet delivery rates in scenarios where nodes are unevenly distributed.