Resource Optimization In Wireless Networks Of UAV Assisted Communications

The rapid maneuvering characteristics and line of sight(LoS)channel advantages of unmanned aerial vehicle(UAV)have attracted extensive attention from academia and industry.Similar to the existing mobile networks,UAV wireless networks also have difficulties and challenges in efficient use of wireless resource,quality of service(QoS)guarantee,physical layer secure communication,and weak user signal enhancement.And how to provide solutions to these problems for U AVassisted wireless networks by combining the rapid deployment capability and airground channel conditions of UAV is the key to improving network performance at present.For addressing the aforementioned problems,this paper focuses on the energy efficiency and sum rate optimization of UAV-assisted wireless network,and develops the deployment design scheme of UAV-based wireless network around the topics of efficient resource allocation in fog UAV network,QoS guarantee of UAV users,secure communication of dual UAV networks,and signal enhancement of UAV users in intelligent reflecting surface(IRS)scene,which reveals the design and optimization mechanism of energy-efficient communication of UAV-assisted wireless network,as shown in the following four aspects.(1)Aiming at solving the difficult problem of the integration of fog radio access network and UAV wireless network,a scheme of caching UAV network deployment based on fog radio access network is proposed,and the user caching and operation mechanism in fog UAV network is designed.Considering the scarcity of spectrum resources in UAV network,an energy efficiency optimization model of UAV downlink network based on non-orthogonal multiple access(NOMA)technology is established.In the process of solving the problem of subchannel allocation in UAV network,a time-varying matching and swapping algorithm for subchannel-user connection based on the flight angle of fog UAV is proposed.At the same time,based on the problem of power allocation in UAV network after subchannel optimization,this work proposes a power optimization scheme in UAV network based on the alternating direction method of multipliers(ADMM)method.Finally,the performance differences between different schemes are verified by simulation experiments.The advantages of fog UAV network architecture and NOMA-enabled UAV downlink design are also proved.(2)Aiming at solving the problem of multi-user QoS guarantee in UAV based wireless network,considering the influence of UAV flying trajectory on air-ground LoS channel,a trajectory representation model of UAV based on two-dimensional discretization is established.At the same time,considering the QoS constraints of UAV users,the energy efficiency maximization optimization problem of UAV wireless network enabled by NOMA is proposed,which is divided into three subproblems of user scheduling,power allocation and flying traj ectory optimization in UAV network.In the solution,a heuristic user scheduling algorithm for UAV network in different time slots is proposed.Then,the power optimization problem of UAV network is transformed into a convex optimization problem by logarithmic approximation theory.The optimal solution is obtained by Lagrange dual method.Finally,the successive convex approximation(SCA)optimization method based on the first-order Taylor expansion is used to optimize the flying trajectory of the UAV.Simulation results show that the performance of the proposed resource optimization algorithm is higher than the existing methods,and the superiority of the three-stage joint optimization process in UAV network is also verified.(3)Aiming at solving the physical layer security problem in UAV network,this paper proposes a cooperative anti-eavesdropping scheme based on the deployment of dual UAVs,which can protect the information of legitimate users through the service from communication UAV and interference from jamming UAV.At the same time,considering the actual estimation error of UAV in the air-to-ground channel,the imperfect channel state information(CSI)based UAV channel model is established,and the optimization problem is modeled as a secure energy efficiency maximization problem with network outage probability constraints.This problem limits the maximum flying energy consumption of the communication UAV under the 3D trajectory,and comprehensively considers the network resource limitations of the dual UAVs.In order to solve the non-convex optimization constraint in the above optimization model,the non-probabilistic transformation of network outage probability requirements is carried out,and the optimization schemes based on wireless resource allocation,flying trajectory of communication UAV,and noise power design of jamming UAV are proposed by using approximate optimization and auxiliary variable scaling.The simulation results show that the deployment of jamming UAV and the trajectory optimization of communication UAV can improve the secure energy efficiency in UAV wireless network.(4)Aiming at solving the signal enhancement problem of users at the edge of UAV network,a UAV communication scheme with multi-IRSs is proposed.Considering that the direct communication link between UAV network and edge users is blocked,multi-IRSs are used to build a signal enhancement channel for weak users,and a sum rate optimization problem in UAV network based on IRSuser association optimization,UAV hovering altitude optimization,and IRS phase shift design is established.Considering the strong coupling relationship between IRS-user association and the decoding order of successive interference cancellation(SIC)in downlink NOMA networks,a decoupling scheme based on co-channel interference value expansion is proposed to transform the problem into a convex optimization problem about the association results.Based on the nonconvex optimization transformation process of SCA,the optimization scheme of UAV hovering altitude is designed.Finally,the gradient approximation univariate optimization algorithm is used to optimize the coefficient of IRS reflection unit to get the phase shift design scheme of IRS.The simulation results show that the proposed algorithm can improve the sum rate of users,and point out that it is positively affected by the number of reflection units.