Roadside units placement within city-scaled area in vehicular ad-hoc networks

We aim through this work to investigate the optimal deployment of roadside units, the stationary component in the vehicular network, in order to improve the performance, the reliability and the connectivity of the network. RSUs have a critical role in safety applications but also in infotainment applications. Hence, spreading RSUs should pay attention to cover with its best effort the dark zones and to serve the biggest number of vehicles with continuous services while remaining in the range of the deployment budget. To reach our objectives, we consider stationary and mobility-based statistics to model our traffic network as a graph with weighted links showing the importance of each link in terms of amount of vehicles, hazardous level, and mobility-based parameters such as average of connected time based on speed information. Then, all the potential positions are designated and two different algorithms, Knapsack and PageRank, are applied to select the optimal set of RSU positions within the cost range. Finally the output set of each algorithm is evaluated against real traffic records and statistics to show the network performance.