Research On Resource Management And Deployment Optimization In UAV-enabled Communications

In recent years,with the rapid development of mobile communication technology and un-manned aerial vehicle(UAV)industry,the applications of UAV in the field of wireless com-munication have become a research hotspot.UAVs can be used as aerial base station(ABS)or aerial mobile relay,which have unique advantages in emergency or temporary communi-cation scenarios.Different from the traditional base station on the ground,UAV base stations have some special advantages,such as fast and flexible on-demand deployment,better line-of-sight(Lo S)communication link,high mobility in 3D space.UAVs can flexibly adjust their locations according to the location distribution and requirement of users on the ground,consequently further enhance the performance of the communication networks.However,despite many advantages,there are many challenges at the same time,such as 3D deploy-ment optimization,resource management,energy efficiency,trajectory optimization,etc.In this thesis,the resource management and deployment optimization problems in UAV-enabled communication networks are studied,where UAVs used as aerial base stations in post-disaster rescue and temporary hotspot areas for both interference-free and interference scenarios.The main works are summarized as follows:(1)The scenario in which multiple UAVs provide service to ground users in a disaster area is considered.In order to avoid interference among UAVs,orthogonal spectrum resources are used among UAVs.The total downlink transmission power of UAVs is minimized through joint optimization of user association,bandwidth allocation and 3D locations of UAVs under the conditions of satisfying the minimum rate of each user and system load balancing con-straints.Since the original problem is hard to solve directly,the original problem is firstly decomposed into three subproblems,and then it is finally solved by alternating iteration algo-rithm.When the UAV’s 3D locations are determined,a modified Kmeans algorithm is used to obtain balanced user clustering.When the user association and UAV’s 3D locations are determined,the convex optimization method is used to obtain the optimal bandwidth alloca-tion.When user association and bandwidth allocation are determined,a modified differential evolution algorithm is proposed to optimize the 3D locations of the UAVs.Simulation results show that under the conditions of satisfying the minimum rate of ground users and system load balancing,the proposed algorithm can effectively reduce the transmission power of UAVs compared with the existing algorithms.(2)The hotspot scenario with limited resources is considered,where multiple UAVs are de-ployed to serve multiple ground users in the same frequency band.Subchannel allocation,power allocation and UAV’s height are optimized jointly to maximize the total user through-put of the network.Since the original problem is difficult to solve directly,it is also decom-posed into several subproblems to solve.Firstly,the horizontal 2D locations of UAVs and user association are obtained by using Kmeans++ algorithm,and then the subchannel allo-cation,power allocation and UAV’s height are optimized jointly.When the user association and UAV’s 3D locations are determined,for each UAV,an improved Hungarian algorithm is proposed to assign subchannels to each user,and then the water-filling algorithm is used to obtain the optimal transmit power allocated on each subchannel.When the resource al-location is determined,the height optimization problem of each UAV is a one-dimensional non-convex optimization problem,the golden section method is used to search for the op-timal height of each UAV.Finally,the optimization is performed between multiple UAVs interactively until convergence.Simulation results show that the proposed algorithm can effectively improve the system throughput compared with other benchmark algorithms.