Service-Throughput Maximization for Air-Ground Cooperative Vehicular Networks

Researchers have proposed employing unmanned aerial vehicles (UAVs) equipped with dedicated short-range communication (DSRC) devices to provide supplemental vehicle-to-infrastructure (V2I) links in temporary traffic hot spots where static roadside units (RSUs) are lacking. However, existing studies have largely ignored the limited communication range of UAVs, which leads to the inefficiency of vehicular networks. To address this limitation and enhance the overall performance of vehicular networks, we propose an innovative air-ground cooperative architecture. The architecture involves the utilization of UAVs equipped with DSRC devices, serving as flying-RSUs, to provide temporary V2I links for vehicles. Simultaneously, we consider vehicle-to-vehicle (V2V) links to extend the coverage and service capabilities of UAVs. Under this air-ground cooperative architecture, we formulate a service-throughput maximization problem by optimizing the 3-dimensional locations of UAVs. The primary goal is to find the optimal locations of UAVs that lead to the maximum number of served vehicles while satisfying their data rate requirements. Since the problem includes a large number of binary and continuous variables, we propose a hybrid firefly and particle swarm optimization algorithm to efficiently solve it. Simulation experiments based on real taxi trajectory data in Shanghai verify the effectiveness of our proposed algorithm.