| So far,both academia and industry have been exploring the 6th generation mobile networks(6G).6G is envisioned to achieve the ubiquitous wireless coverage for a space-air-ground integrated network a ubiquitous wireless coverage for space-air-ground integrated networks,and satisfy the timely and reliable wireless connections for various devices.Due to the high flexibility and high cost-effectiveness,the unmanned aerial vehicle(UAV)-based communication networks have been considered as an effective complement to the existing ground communication networks.The UAV-based communication is a promising technique for various fields,e.g.,live broadcast,protection,rescue,etc.Existing researches on UAV-based communication networks mainly concerns the single-UAV-based network.However,complex scenarios,such as massive data collection and space-ground integration,have higher demands on coverage range and communication capabilities,which can be met by employing multi-UAV-based communication networks.Hence,researches on multi-UAV-based communication networks are gaining huge momentum,while facing many challenges,especially the optimization of energy consumption.The flexibility of a multi-UAVbased communication network consists of network topology,and the network connections,quality of services(QoS),and coverage range,that need to be jointly considered for higher efficiency.Therefore,to fully utilize the spatial and wireless resources,an energy-efficient joint design of deployment and wireless resource allocation is in dire need.The energy consumption optimization of a multi-UAV-based communication network is crucial and needs to be further studied.In this study,we study the problems of large-scale wireless transmission,the trade-off between energy stability and transmission rate,and the coverage for mobile users(MUs).First,we consider multi-hop UAV communication for large-scale wireless transmission.The limited number of UAVs and transmission power affect the network connection and link capacities,and thereby affecting the energy consumption and end-to-end throughput.Second,we consider solar-powered UAVs to prolong the endurance.Due to the instability of the solar supply and wireless link status,the transmission rate of each link and long-term energy consumption are unstable.Third,multiple UAVs need to cooperatively provide wireless coverage when ground MUs’ demand and location are random.In such cases,QoS,transmission power,and residual energy of UAVs are different,and thereby the energy efficiency,charge scheduling sequences,and endurance.The main contributions of this paper are as follows:1.To achieve the large-scale transmission,we employ the multi-hop UAV communication,and propose an energy-efficient scheme of joint UAV deployment and wireless resource allocation.First,considering the constraints of transmission rate requirements,network connections,the maximum transmission power,and flight position limits,we formulate a transmission power minimization problem of joint three-dimensional(3D)placement,multi-hop routing,transmission rate and power control.Second,due to the high complexity of the formulated problem,which cannot be directly solved,we reformulate it into two subproblems,i.e.,a subproblem of routing,power,and transmission rate allocation,and a subproblem of 3D placement of UAVs.Third,we design an errorbounded linearization approximation algorithm to solve the first subproblem,and derive the maximum theoretical performance gap.To solve the 3D placement subproblem,we transform it as a global consensus problem,and solve it by our proposed alternating direction method of multiplier-based distributed algorithm.In particular,such distributed algorithm allows UAVs to optimize their location in parallel,which significantly reduces the implementation complexity of the proposed algorithm.Finally,the extensive simulation results prove the effectiveness of our proposed scheme.The energy efficiency can also be improved by using our proposed scheme.2.To achieve the balance between the stability of the energy system and transmission rate,we propose a long-term energy-efficient dynamic resource allocation scheme.First,the solar powered-UAVs are employed to prolong endurance.Due to the instability of solar supply and the wireless link status,we consider the solar power and channel gain as random variables following the independent and identically distributed process.Second,considering the constraints of system stability,real-time transmission rate requirements,and the maximum transmission power,we formulate a long-term energy consumption minimization problem of joint routing,power control,and transmission rate control.Third,we employ the Lyapunov theorem to transform the formulated problem as an online optimization problem,which can be solved without prior knowledge.Then,we design a decomposition-based distributed algorithm to solve the online problem and prove the optimality and stability guarantee of the proposed algorithm.Finally,the extensive simulation results prove that the proposed scheme can improve the long-term energy efficiency as well as guarantee the system stability.3.To achieve the cooperative coverage for the MUs,we propose a deep reinforcement learning(DRL)-based scheme of joint dynamic UAV scheduling and wireless resource allocation.First,considering the constraints of maximum transmission power,location and residual energy of UAVs,and QoS requirements of MUs,we formulate an energy efficiency maximization problem of jointly optimizing the real-time 3D position of UAVs and power control.Second,we reformulate the original problem as a Markov Decision Process(MDP).In the formulated MDP,we optimize UAVs’ position and transmission power based on the UAVs’position and residual energy,as well as MUs’ position and QoS requirements,are considered.Then,we design a DRL-based algorithm to solve the MDP problem.Finally,the extensive simulation results illustrate the convergence performance of the proposed training algorithm,the adaptation to the dynamics,and the improvement of energy efficiency achieved by our proposed scheme. |
PDF Full Text Request