Research On Irregular Intelligent Surface-Assisted Cell-Free Massive MIMO Systems

Recently,the cell-free massive multiple-input multiple-output(MIMO)and reconfigurable intelligent surface(RIS)are extensively studied as the most promising technologies of the 6G wireless communication.On the one hand,the user-centric cell-free massive MIMO system eliminates the concept of cell and inter-cell interference.On the other hand,RIS with low energy consumption and cost,which consists of a large number of reflecting elements,can reconfigure propagation channels between access points(APs)and users via adjusting the phase of each element.These aforementioned advantages make the RIS-aided cell free massive MIMO system attractive.However,the system channel estimation overhead arises for purely a great deal of RISs cannot be ignored,which lead to increase the system capacity with a reasonable number of RISs becomes a pressing issue.In order to solve this problem,this work introduces the concept of irregular RIS,and mainly studies the downlink communication process of cell-free massive MIMO system assisted by irregular RIS.The main contributions of this dissertation are summarized as follows.For the problem of maximizing the weighted sum rate(WSR)of users in the irregular RIS-assisted cell-free massive MIMO system,this work considers these three constraints including the limited transmit power of the AP,the unit mode limit of the RIS phase shift element,and Irregular RIS discrete topology and proposes an AP-RIS joint alternating optimization scheme based on quadratic transform(QT).In this scheme,we utilize QT and dual transformation techniques to maximize the WSR problem and decouple the transformed equivalent non-convex problem into three convex subproblems: AP active beamforming,RIS passive phase shift and irregular RIS topology,followed by alternate optimization to obtain a suboptimal solution to the original problem.Two of the beamforming subproblems were similarly transformed with the help of QT techniques.The simulation results show that the irregular RIS can greatly improve the capacity and user scalability of the cell-free massive MIMO system.For the problem of maximizing the minimum achievable rate among users in the irregular RISassisted dell-free massive MIMO system,this work considers two cases of RIS continuous phase shift and discrete phase shift,and proposes an AP-RIS joint alternating optimization scheme based on the maximum-minimum fairness criterion.For the case of continuous phase shift,after the two subproblems of AP active precoding and RIS passive phase shift having relaxed into a convex feasibility problem,the suboptimal solutions can be obtained effectively by binary search.In the case of discrete phase shifts,the AP uses active zero-forcing beamforming to iteratively obtain suboptimal solutions for all RIS phase shift elements through a successive refinement algorithm.The simulation results demonstrate that the performance of the irregular RIS architecture is significantly better than that of the conventional RIS architecture.In addition,it also proves that our proposed AP-RIS joint alternating optimization scheme is superior to the existing optimization schemes.Finally,this work demonstrates that the phase shift setting with 2-bit quantization resolution is only slightly worse than the ideal continuous phase shift setting,which allows us to save deployment cost and computational overhead in practical applications.