The Method On Energy-efficient UAV Scheduling And Deployment Planning In Wireless Networks

With the development of 5G/6G,the requirements of users for system communication performance continue to increase.While wireless networks that can be seen everywhere bring convenience to people,they also encounter system performance bottlenecks.According to different types of wireless networks,this thesis divides wireless networks into wireless rechargeable sensor networks(WRSNs),wireless powered communication networks(WPCNs),device-to-device(D2D)networks and reconfigurable intelligent surface(RIS)-assisted wireless cellular networks.Among them,WRSNs are emerging networks under the 5G/6G architecture,while WPCNs can be regarded as WRSNs with data transmission functions.Compared with traditional networks,these two networks can greatly improve the energy efficiency of the system.As two emerging technologies,D2 D and RIS can further promote the energy efficiency of the networks.However,these wireless networks all face performance bottlenecks.For this purpose,this thesis considers the UAV deployment planning and trajectory scheduling in WRSNs,WPCNs,D2 D networks and RIS-assisted wireless cellular networks to maximize the energy efficiency of the system while ensuring network communication performance.The main contributions and innovations are as follows:1.Joint optimization of a single UAV trajectory scheduling and energy efficiency(1)A joint-charging unmanned aerial vehicle(CUAV)scheduling and trajectory optimization problem(JSTOP)is formulated to simultaneously minimize the number of CUAV hovering point,the number of repeatedly charged sensor nodes,and the flight distance of the CUAV for charging all sensor nodes.The originally formulated JSTOP is divided into two sub-problems,which are CUAV scheduling optimization problem(CSOP)and CUAV trajectory optimization problem(CTOP).Therefore,two different improved versions of the particle swarm optimization(PSO)algorithms,which are PSO with a flexible dimension mechanism,a K-means operator and a punishment-compensation mechanism(PSOFKP),and PSO with a discretization factor,a 2-opt operator and a path crossover reduction mechanism(PSOD2P),are proposed to solve the converted CSOP and CTOP,respectively.Simulations evaluate the effectiveness and performance of the proposed CUAV-based charging method,and the proposed algorithms are superior in solving the formulated CSOP and CTOP under different scales and network settings.In addition,the stability of the proposed algorithms is also verified.(2)Based on the abovementioned research,this thesis further extends the flight space of UAV to the three-dimension(3D)space with obstacles,and the WRSNs are extended to the WPCNs.To this end,a joint-UAV power and 3D trajectory optimization problem(JUPTTOP)is formulated to simultaneously increase the total number of covered wireless devices,improve time efficiency and reduce the total flight distance of the UAV.In order to solve JUPTTOP,the formulated JUPTTOP is divided into two sub-optimization problems,which are the UAV power allocation optimization problem(UPAOP)and the UAV 3D trajectory optimization problem(UTTOP).For UPAOP,an improved non-dominated sorting genetic algorithm-II with a K-means initialization operator and a Variable dimension mechanism(NSGA-II-KV)is proposed to solve it.For UTTOP,a pretreated method is introduced,and then an improved particle swarm optimization with a normal distribution initialization,a genetic mechanism,a differential mechanism and a pursuit operator(PSO-NGDP)is used to solve this problem.Simulation results verify that the proposed algorithms can effectively solve the converted UPAOP and UTTOP,respectively.2.Joint optimization of multi-UAV deployment planning and energy efficiency(1)In a UAV-assisted D2 D network,the number of deployed UAVs,UAV locations,UAV transmission powers,UAV flight velocities,communication channels and UAV-device pair allocation jointly affect the D2 D network capacity.In order to simultaneously consider system energy efficiency,a UAV-assisted D2 D network resource scheduling optimization problem(Net Res SOP)is formulated,jointly considering all the abovementioned decision variables to maximize the D2 D network capacity,minimize the number of deployed UAVs,and minimize the average energy consumption of all UAVs.Therefore,a non-dominated sorting genetic algorithmIII with a flexible solution dimension mechanism,a discrete part processing mechanism,and a UAV number adjustment mechanism(NSGA-III-FDU)is proposed to optimize all optimization objectives simultaneously.Through simulations,it is verified that the proposed NSGA-III-FDU can effectively solve Net Res SOP under different scales and settings.(2)Based on the abovementioned research,this thesis further considers the combined application of UAV and RIS,and researches on UAV-mounted RIS-(UAV-RIS)assisted wireless cellular networks.First,an energy-efficient communication problem based on multi-objective optimization framework(EEComm-MOF)is formulated to jointly consider the beamforming vector of base station,the location deployment and the phase shifts of UAV-RIS system so as to simultaneously maximize the minimum available rate,maximize the total available rate of all ground users,and minimize the total energy consumption of the system,while the transmit power constraint of base station and the discrete phase shifts of UAV-RISs are considered.Second,an improved non-dominated sorting genetic algorithm-II with continuous solution processing mechanism,discrete solution processing mechanism,and complex solution processing mechanism(INSGA-II-CDC)is proposed.Finally,simulations results demonstrate that the proposed INSGA-II-CDC can solve EEComm-MOF effectively and outperforms other benchmarks under different parameter settings.