Research On Uav-aided Communications In Vehicular Networks

As a typical application scenario of Internet of things,vehicular networks have raised extensive research interests in both the academic community and communication industry in recent years.With the ever-increasing demand for service types and user traffic in vehicular networks and the continuous advancement of unmanned aerial vehicle(UAV)technology,introducing UAV into the vehicular networks as a promising communication solution for assisting communication in vehicular networks can alleviate the pressure on the communication infrastructure and further improve the network capacity.However,most of the existing studies on UAV deployment in vehicular networks mainly focus on how the UAVs serve the vehicle users,i.e.,the UAV-to-Vehicle(U2V)communications,neglecting the utilization of vehicle-to-vehicle(V2V)communications in the networks.Even when it comes to studies on V2V communications,the influence of vehicle’s high mobility characteristics on communication services performance is often not fully considered.In addition,researches on UAV-assisted vehicular communication systems lack theoretical analysis from a statistical perspective and the optimization of UAV deployment based on this global network performance analysis,which makes the network services capacity is not fully exploited.To address the above problems,this thesis focuses on the UAV deployment to assist vehicular networks communications during peak-traffic hours.The study combines UAV transmission of V2V transmission and solves the optimal UAV deployment problem based on the analysis of the successful service performance of vehicle users in the network.The main research contents are as follows.(1)The UAV altitude deployment in a vehicle user-concentrated single hotspot area.We first derive the successful service probability for the U2V transmission.Meanwhile,taking those important factors,i.e.,vehicle mobility and social proximity,into account,we estimate the successful service probability for the V2V transmission.The average successful service probability through the U2V transmission and V2V transmission for the considered scenario is then derived.Based on the mathematical analysis results,the service probability maximization problem is formulated and the system performance is improved by optimizing the UAV’s altitude position.To find the optimal solution,a particle swarm optimization algorithm is proposed.Finally,we solve the optimal altitude in simulation and numerical results has verified the theoretical analysis and the efficiency of our proposed scheme.(2)The UAV three-dimension position deployment in the multi-hotspot area with time evolution.We first model and describe the dynamic distribution of the multi-hotspot area in the considered scenario and the UAV’s energy consumption.Then the vehicle users’successful service probability for U2V transmission,V2V transmission and the average successful service probability in a single time slot is fully analyzed.After that,based on the above mathematical analysis and taking UAV’s energy consumption constraint into consideration,the real-time UAV position deployment problem during the peak-traffic duration is formulated as the average number of successful service users maximization problem.And a complete design scheme for optimal UAV position deployment is proposed based on reinforcement learning methods.