Research On Key Technologies Of Intelligent Reflecting Surface-aided Millimeter Wave Communication Systems

With the rapid development of wireless communication technologies,wireless data services have also grown explosively.In order to meet the needs of greater bandwidth and higher spectrum efficiency of future wireless networks,the millimeter-wave(mm Wave)technology has been attracting much attention.Although there are abundant spectrum resources in mm Wave band,the key problem for mm Wave communication is its serious path loss.One way to overcome this issue is to deploy more base stations,which will obviously cause high hardware costs.The recently developed intelligent reflecting surface(IRS),a nearly passive equipment,can be deployed in wireless communication systems to effectively improve the quality of wireless channels and can also be used to enhance the transmission security.In order to boost the development of IRS-aided mm Wave communications in different scenarios,this thesis studies the sum-rate and transmit power optimization issues in IRS-aided mm Wave communication systems.Firstly,this thesis considers a mm Wave multiuser multiple input single output(MISO)downlink communication system with the assistance of multiple IRSs.The objective is to maximize the sum-rate of all users by jointly optimizing the active transmit beamforming and passive reflection phase shifts.To deal with the coupling objective function,the alternating iterative algorithm is leveraged to decouple the active and passive beamforming design.Then,the closed fractional programming method is applied to deal with the non-convex problem.As for optimizing the transmit beamforming,we introduce two auxiliary variables to optimize it iteratively.In order to optimize the reflection phase shifts with non-convex constraints,we adopt a gradient descent method,and the inexact one-dimensional step size search method is applied to update the step size.The simulation results demonstrate that the optimization of active and passive beamforming can significantly improve the quality of the considered communication system.Non-Orthogonal Multiple Access(NOMA)is a promising technology in B5 G due to its high spectral efficiency and the capability of supporting massive connectivity.This thesis studies an IRS-aided mm Wave power domain-NOMA(PD-NOMA)downlink transmission system.Our objective is to minimize the transmit power of the base station by alternately optimizing the transmit beamforming and reflection phase shifts,subject to the secure communication and the quality of service(Qo S)constraints of users.This thesis proposes an alternative optimization algorithm and the semidefinite programming relaxation to optimize the active beamforming and passive beamforming vectors.Gaussian randomization and eigenvector corresponding to the largest eigenvalue methods are used to solve the rank 1 constraint.The simulation results demonstrate that the received power at the eavesdropper can be effectively reduced by optimizing the active and passive beamforming,and hence the security of the system can be ensured.Furthermore,the influences of the number of reflection units,the minimum Qo S requirements of each user,and the quality of channels on the minimum transmit power are carefully analyzed.