Research On Performance Of Unmanned Aerial Vehicle Communication System Based On Non-orthogonal Multiple Access

Conventional base station(BS)communications are readily affected by the obtacles,where the path-loss is strong,which leads to the small coverage area.By carrying the base station(BS)on the unmanned aerial vehicles(UAVs),UAV communication is capable of providing access services to ground users or providing relay services to other UAVBSs.UAV communications have the characteristics of flexibility and mobility,which is one of the most critical techniques for improving communications quality.This dissertation aiming at the limited energy on-board of the UAV-BSs,integrating non-orthogonal multiple access(NOMA)technique and UAV communications,provides the researches of UAV communication system models,user pairing strategies,as well as user's performance analysis.In order to improve the spectral efficiency and energy efficiency of the UAV communication systems,by utilizing order statistic and stochastic geometry techniques,this dissertation proposes several NOMA-UAV communications models,and the network's performances have been researched with the different communication scenarios,the different channel status,the different number of antennas,as well as the different number of users.The main innovative contribution of this dissertation can be concluded as follows:(1)Aiming at the problem of limited energy on-board for UAV communications,an air-to-ground(A2G)communication model based on NOMA is established,and the performance analysis of UAV A2 G communication system based on order statistics is proposed.According to the statistical user pairing strategy,by utilizing the exponential expansion approximation and the binomial coefficient approximation,the performance of the outage probability of individual users,the outage probability of sum rate,and the outage probability of the system in A2 G communication systems are analyzed.The diversity order of individual users is related to the product of the user index number and the channel fading coefficient.The diversity order of sum rate is related to the product of the index number and the channel fading coefficient of the user with stronger channel gain.The proposed method is capable of exactly analysing the performance of the UAV A2 G communication networks.In order to solve the high-complexity of user-pairing strategy based on order statistics,a performance analysis method of UAV A2 G communication system based on stochastic geometry is proposed.According to the geometric user pairing strategy,by utilizing the signal alignment technique,the small-scale fading matrix of the paired users are identical,we design a precoding matrix and detection vector based on the zero-forcing coding.By using the Box-Muller transform and the Riemann-Stieltjes integration method,the network's performances are analyzed,e.g.,the outage probability and ergodic rate of the ground users.Both of the analytical and simulation results illustrate that in the case of high signal-to-noise(SNR)ratio,the diversity gains of the ground users,which are all1,has no relation to the channel fading coefficient and the number of antennas.The highSNR slopes of near users and far users are 1 and 0 respectively,which can be used for approximating the performance of the UAV A2 G communication systems.illustrate that in the cases of different channel fading coefficients,the diversity gains of the ground users are 1;the ceiling of ergodic rates of far users occurs,and hence the high-SNR slopes are0.On the contrary,the ergodic rates of near users increase versus the SNR,and the high SNR slopes of near users are 1.(2)Aiming at that the user association strategy of the conventional cellular networks is not applicable for the UAV A2 G cellular networks,and hence an A2 G cellular network communication model based on NOMA is established.The performance analysis methods of the A2 G cellular emergency communication system and offloading communication system are proposed based on stochastic geometry.By utilizing the multitier homogeneous Poisson point process(HPPP),Poisson cluster process(PCP)and Poisson hole process(PHP),the location randomness of UAV-BS and ground users can be analyzed.Based on the principle of the nearest instantaneous distance,a user association strategy based on geometry is designed.By using Faa di Bruno's theorem,we analyze the coverage probability performance of the ground users in the proposed A2 G cellular networks in the case of imperfect successive interference cancellation(IP-SIC).In the case of the low IP-SIC coefficient,NOMA-UAV cellular networks can work normally,and the performance of the UAV A2 G cellular network communication systems can be approximately analyzed.The simulation results show that the offloading cellular network is affected by IP-SIC dramatically,where the coverage probabilities of users are all 0 at the high SNR regime in the case that IP-SIC coefficient is 0.5,whereas the coverage probabilities of users are nearly identical of the emergency communication networks.(3)Aiming at the high complixity of the performance analysis of UAV integratedair-ground(IAG)networks,an IAG communication network based on UAV-BS is established,and a performance analysis method is proposed for the proposed network.By utilizing the three-dimension(3D)binomial point process(BPP),a geometry-based user association strategy is proposed.By using the Taylor expansion approximation of the lower incomplete Gamma function,the performances,e.g.,outage probability and ergodic rate,of the IAG networks in different fading channels are analyzed.Simulation results confirm the correctness of the proposed analysis method in the traditional fading channels,and illustrate that the PLo S channel can be approximated to the channel with the lowest channel fading coefficient.(4)Aiming at the scarcity of spectral resources cannot satisfy the increasing number of UAV and ground users,an IAG communication network based on the ground BS and its performance analysis method are established.By utilizing 2D-BPP and 3-BPP simultaneously,a user pairing strategy based on geometry is proposed,and hence any nember of NOMA-users can share the same spectral resource.By using the Taylor expansion approximation of the Gaussian hypergeometric function and Gauss-Chebyshev integral,the coverage probabilities of the ground users and aerial users in the proposed networks are analyzed.It is shown that since the SIC technique is employed at the ground BS,error propagation occurs.The simulation results confirm the effects of error propagation.