Optimization Of Task Allocation In UAV-based Wireless Communication

Due to the characteristics of fast speed,flexible deployment location,excellent channel conditions and low cost,Unmanned Aerial Vehicle(UAV)has unparalleled advantages in the field of wireless communication.In recent years,due to the gradual maturity of the UAV industry,its application cost has gradually decreased,and has been gradually used in various industries such as photography,Internet of Things control,express transportation,disaster relief,etc.The use of UAV-assisted wireless communication has also become a current research hotspot.Although the advantages of UAVs are obvious,its many limitations limit its wider application.The most important point is the limited battery life of UAVs.Due to the use of battery power,UAV cannot complete long-term continuous work.Therefore,minimizing the energy consumption of the UAV during its work will effectively increase the working time of the UAV,thereby providing longer service to adapt requirements of various scenarios.In addition,when the communication scale is large,no matter how efficient the energy optimization algorithm is,a single UAV cannot meet the communication requirements in a single flight.For this reason,it is necessary to divide the communication tasks appropriately,then finish them by using multiple flying times or dispatching multiple UAVs in order to complete communication tasks in largescale scenarios.In response to the challenges in the above two UAV application scenarios,this article has made two contributions:First,ignore the upper limit of the airborne power of the UAV,the relationship between the energy consumption of the rotary-wing UAV and the communication strategy and flight trajectory is established under the two-dimensional distribution and communication requirements of a specific ground terminal model,which is expressed as a non-convex function with multiple variables.And by discretizing the flight trajectory and successive convex approximation technology,the original problem is transformed into a convex optimization problem for multiple iterations,so as to obtain an approximate optimal solution that can meet the conditions.In addition,considering the limited airborne power of UAVs,this paper further studies the appropriate partition method of all tasks when there are more ground terminals and a large amount of communication data.By using an adaptive clustering algorithm based on dividing and merging,all tasks are divided into as few sub-tasks as possible,and each sub-task can be completed in one flight of the UAV.Therefore,when it is inevitable to perform multiple flights or dispatch multiple UAVs to complete the task,the algorithm is able to reduce the number of single UAV flying times or the amount of dispatched multiple UAVs to reduce costs and improve efficiency.In order to verify the performance of the algorithm,this paper designs multiple sets of comparative experiments for the two algorithms.For the energy minimization algorithm,we used time minimization,hovering in the center of all ground terminals and flying along the shortest flight path.Comparing with these three flight strategies,we proved the advantages of the algorithm proposed in this article in energy consumption;For clustering algorithm,this paper uses the adaptive K-Means algorithm for comparison,which proves the optimization of the clustering algorithm proposed in this paper on the total number of clusters.