The Research Of Unmanned Aerial Vehicle Networks Based On SDN

Unmanned Aerial Vehicles(UAVs)can construct flexible airborne network and provide communication service to ground users due to their flying agility,which shows great prospect for disaster relief,military operation and so on.However,there are still some challenges in UAV networks caused by the mobility and energy constraint of UAVs,the resource management in the networks,and the differences in higher layer protocols.As a new type of network architecture,Software Defined Networking(SDN)decouples the control logic from the data forwarding devices and provides a way to programmatically control networks.The programmability and logically centralized control ability of SDN can manage the networks and solve the above problems in UAV networks.This research focuses on the UAV networks based on SDN and the main contents are as follows:We first introduce the related knowledge about UAV networks and software defined networking and analyze the star topology and mesh topology in UAV networks.Then,we categorize the UAV networks based on different using purposes and also study the architecture of SDN and OpenFlow technique.Furthermore,the centralized control mode and distributed control mode in UAV networks based on SDN are discussed and the advantages and drawbacks of both modes are proposed in terms of deployment speed,reliability,flexibility and so on.In UAV networks based on the centralized SDN control,a real-time monitoring platform is proposed to effectively monitor and manage the network resource in UAV networks.The monitoring platform can analyze the data information from UAV networks and provide the reliable basis for decision making and algorithm deployment.We first analyze the key issues for designing the monitoring platform,then the architecture of the monitoring platform and its main function modules are proposed.At last,simulations validate the function of our proposed monitoring platform including the management of both UAV networks and user equipment and the monitoring of link load and packet loss rate.Furthermore,we study the distributed SDN control architecture for the UAV networks and apply the monitoring platform in the distributed controllers.The messages delivered between controllers are defined to guarantee that the distributed controllers in UAVs can acquire the network topology and the information of traffic and users from other UAVs.Based on the proposed network architecture,a preplanning algorithm is proposed.The battery power of UAVs,network load and the priorities of data flows are considered in our algorithm.When the monitoring platform detects the battery power of the UAV exceeds the threshold,the algorithm can plan the routing path for the flows that pass through the UAV and switch the link fast.The proposed algorithm solves the communication interruption problem caused by the UAVs that are out of power.At last,we construct our simulation environment.The messages delivered between distributed controllers in UAVs are implemented through the simulations and the simulations also validate characteristics and validity of our proposed algorithm.The results show that the optimal routing path is selected to switch in our proposed algorithm which guarantees the continuous communication between users.