Research On UAV-Assisted Wireless Communication Network Technology

With the continuous development of wireless communication networks and unmanned aerial vehicle(UAV)based technologies,UAVs have gradually become a hot research topic in the field of wireless communications.UAVs can enhance terrestrial wireless communication networks by expanding their coverage,improving capacity,providing emergency communications and flexible networking,etc.,as a result,UAVs have become promising potentials for the development of next-generation networks.However,UAV-assisted wireless communication networks also face many challenges.How to improve resource utilization and energy efficiency,ensure low latency,and implement effective mobility management is particularly important.This dissertation introduces the large-dimensional random matrix theory,convex optimization tool and considers the issues of the spectrum resource allocation,the deployment position of the UAV,the transmission power,the hardware impairment,the backhaul compression,the channel estimation error,etc.,in the UAV-assisted wireless communication networks.The system performance,such as the uplink and downlink sum-rate,is studied,and the impacts of these issues on the system performance are analyzed.The main works of this dissertation are summarized as follows.Firstly,for the downlink scenario in the multi-antenna UAV-assisted terrestrial cellular communication network,the impacts of the spectrum resource allocation and the UAV location deployment on the system performance are analyzed.The UAV communicates with the ground base station through the in-band wireless backhaul link,which shares spectrum resources with the access links of the user equipments(UEs).A maximization problem of the downlink ergodic sum-rate of the UEs is formulated.A low-complexity algorithm is proposed to jointly optimize the spectrum resource allocation factor,the deployment position and the transmission power matrix of the UAV.Specifically,the large-dimensional random matrix theory is utilized to derive the approximate expressions for the downlink ergodic sum-rate of the UEs and the backhaul capacity of the UAV.Subsequently,based on the deterministic approximation expressions,an approximation problem of the original joint optimization problem is established,and,an iterative coordinate algorithm is proposed to obtain the optimal solution of the approximation problem.The simulation results demonstrate the accuracy of the approximate expressions and the effectiveness of the proposed optimization algorithm.Secondly,for the uplink scenario in the multi-antenna UAV-assisted cloud radio access network(C-RAN),the impact of the spectrum resource allocation on the system performance is analyzed.The UAV assists the remote radio heads(RRHs)to collect user data and transmits it to the baseband unit(BBU)pool.The hybrid backhaul link scheme is considered,where wired fiber backhaul links are used among the gateway base station,RRHs and BBU pool,while the UAV is connected to the gateway base station through a wireless backhaul link,which shares spectrum resources with the radio access network.A maximization problem on the uplink ergodic sum-rate is formulated through optimizing the spectrum resource allocation between the wireless backhaul of the UAV and the radio access network.An approximation problem of the original optimization problem is proposed by means of the large-dimensional random matrix theory.As a consequence,the computational complexity and communication overhead are reduced.The simulation results demonstrate the accuracy of the approximate expressions as well as the effectiveness of the optimal solution to the approximation problem.Finally,for the downlink C-RAN scenario assisted by multiple multi-antenna UAVs,the impacts of phase noise,backhaul compression,and channel estimation error on the system performance are analyzed.The UAVs serving as RRHs in the air to provide services to users coordinated by the central BBU pool on the cloud server.The baseband signals are performed regularized zero-forcing precoding and point-to-point compression processing in the BBU pool,before being transmitted to the UAVs through capacity-limited backhaul links,which result in the additional compressive quantization noise.The imperfect channel state information is introduced in the BBU pool during the channel estimation period and the phase noise caused by the non-ideal local oscillators is considered.Subsequently,the approximate expression of the downlink ergodic sum-rate is derived utilizing the large-dimensional random matrix theory.The simulation results demonstrate the accuracy of the approximate expression,and the impacts of phase noise and backhaul compression on the system performance can be theoretically analyzed based on the approximate expression.