Research On 3D Beamforming Based On Intelligent Reflecting Surface

The wireless electromagnetic environment is one of major limitations for improving the performance of wireless communications,which significantly affects the quality of the received signal.In order to solve this problem,innovative technologies are urgently required to overcome the effects of the wireless channel.Recently,as the important candidates for next generation of mobile networks,intelligent reflecting surface(IRS)and 3D beamforming were widely regarded as two promising solutions to improve electromagnetic environment of the wireless communications.Therefore,in this thesis,to meet higher standard of requirements of future wireless communications,the IRS aided 3D beamforming is investigated.The contents are detailed as follows:(1)In IRS aided single user multiple input single output(MISO)communications,maximization of its downlink spectral efficiency is a joint optimization problem of three variables,namely phase-shift matrix of IRS,tilt angle and beamforming vector used in base station(BS)3D beamforming.Based on specific optimization order and bisection search,we first solve this problem by employing a single user alternating optimization(SU-AO)algorithm.But,in each iteration,a specific optimization order of firstly phase-shift matrix,secondly tilt angle and finally beamforming vector is proposed,which facilitates the search of optimal tilt angle in the way of narrowing its trust region and enabling unimodal property over the narrowed trust region.Then,based on the rank one property of BS-IRS channel,a lowcomplexity single user three stage optimization(SU-TSO)algorithm is proposed,which optimizes the phase-shift matrix,tilt angle and beamforming vector by using three stage signal processing scheme.The simulation results show that the proposed SU-AO algorithm and SUTSO algorithm achieve excellent spectral efficiency,and also comfirm that our proposal can solve the limitations of the existed works.(2)In IRS aided multi-user MISO communications,the joint optimization of 3D beamforming and IRS passive beamforming is formulated as a weighted sum rate maximization problem.To solve the optimization problem,based on projected gradient descend and weighted sum mean square error minimization,a multi-user alternating optimization(MU-AO)algorithm is firstly proposed,which provide a universal joint beamforming scheme by alternatively optimizing two sub-problems,namely the joint design of BS’s tilt angle and beamforming matrix,IRS’s phase-shift matrix optimization.Then,based on the high signal-to-noise ratio(SNR)approximation,a simply multi-user three stage optimization(MU-TSO)algorithm is further proposed,its solution is sub-optimally obtained by successively optimizing the phase shift matrix of IRS,tilt angle and beamforming matrix of BS,which can significantly reduce the computational complexity.Finally,a multi-user direct link based optimization(MU-DLBO)algorithm is proposed based on predefined strategy of BS antenna tilt angle.We first fix the tilt angle,and then jointly optimize the beamforming matrix and phase-shift matrix.Accordingly,a better balance of complexity and performance is realized.The simulation results show that our proposed MU-AO algorithm,MU-TSO algorithm and MU-DLBO algorithm can significantly improve the system weighted sum rate performance,which validates the superiority of our proposed schemes.