Research On Multi-domain Resource Optimization Technology For Reconfigurable Intelligent Surface Assisted Cellular Networks

With the rapid increase of wireless mobile service requirements,existing cellular mobile networks confront great challenges in terms of communication capacity,coverage ability and energy efficiency.The first five-generation mobile communication network design treats the wireless propagation environment as fixed,and enhances the network performance through algorithm improvement,hardware upgrade and communication resource increment.However,it is still difficult to resolve the aforementioned technical challenges of future mobile communication networks completely.In recent years,the emergence of reconfigurable intelligent surface（RIS）makes the wireless propagation environment can be reconstructed intelligently.RIS consists of a large number of low-cost passive reflecting elements,which are able to adjust the phase and amplitude of the incident electromagnetic wave.Therefore,designing the RIS’s passive beamforming can achieve various objections such as coverage expansion,useful signals enhancement and interference suppression,reconstructing the wireless environment intelligently.In this thesis,the multi-domain resource optimization technology of cellular network assisted by RIS is thoroughly studied in two scenarios: RIS is deployed on the ground and on the aerial platform,respectively.RIS deployed on the ground is called terrestrial reconfigurable intelligent surface（TRIS）.For this scenario,this thesis optimizes the multi-domain resource optimization allocation in a TRIS-assisted underlaying device-to-device communication network.The spectrum efficiency and energy efficiency of this network are maximized,respectively,by jointly optimizing the user pairing,user transmission power together with active-andpassive beamforming of base station and RIS.Firstly,aiming at the frequency multiplexing problem,this thesis proposes a user pairing scheme based on relative channel strength.Compared with the ideal pairing scheme based on exhaustive search,the performance of this scheme is slightly degraded,but the computational complexity is greatly reduced.Secondly,Lagrange dual transform,quadratic transform and Dinkelbach’s algorithms are exploited to optimize other variables iteratively,and the convergence and complexity of the solving algorithm are analyzed.In addition,this thesis explores the quantitative relationship between RIS’s power consumption and the number of reflection elements together with quantization bits,and establishes a practical power consumption model of RIS.Simulation results show that compared with the traditional D2 D communication network,both the spectrum efficiency and the energy efficiency performance of the proposed TRIS-assisted D2 D communication network are significantly improved.RIS deployed on the aerial platform is called aerial reconfigurable intelligent surface（ARIS）.For this scenario,this thesis studies the optimal multi-domain resource allocation in an ARIS-assisted cellular communication network,in which the base station communicates continuously with the users in a dynamic time-division multiple access（TDMA）manner during the flight.The energy efficiency of this network is maximized,by jointly optimizing the UAV trajectory,user access indicator factor together with activeand-passive beamforming of base station and RIS.Note that this thesis exploits the the practical phase shift model of RIS,in which the reflection amplitude and phase are coupled.Since there is a complex nonlinear relationship between the reflection amplitude and phase in the RIS’s practical phase shift model,this thesis first analyses the interval of the optimal phase,and then proposes a quadratic curve fitting based trust region method to obtain the optimal phase value.Secondly,this thesis comprehensively uses branch-andbound,maximum ratio transmission and Dinkelbach’s algorithm to optimize the remaining variables alternately,and the convergence and complexity of the solving algorithm are analyzed.Simulation results show that compared with the communication-and-hover benchmark,the proposed continuously dynamic-TDMA communication scheme can significantly improve the network’s energy efficiency.