Performance Analysis And Optimization Of Cell-Free Massive MIMO Systems Based On Spatial Correlation

With the large-scale commercialization of the 5th generation(5G)mobile communication system,many countries,research institutions and enterprises around the world have started the research on the 6th generation(6G)mobile communication system.Massive multiple-input multiple-output(MIMO)as one of the key technologies of the 5G can sharply improve the capacity of the mobile communication systems.However,the existing architecture of the mobile communication systems is based on the traditional cells.With the densification of the communication networks,the inter-cell interference inherent in the cell-centric design has become the major bottleneck of the system performance growth,and it is difficult to meet the continuous development demands of the mobile users in the future including the super-large flow,the ultra-low latency,the super-large connection and ultra-high reliability.Cell-Free massive MIMO that utilizes the user-centric design to provide uniform coverage through massive distributed antennas becomes one of the feasible technologies of the 6G.However,the channel spatial correlation between antennas is often neglected in the existing researches,which makes the analysis results inaccurate.In this thesis,the basic theories and key techniques of the cell-free massive MIMO systems based on channel spatial correlation are studied.First of all,this thesis studies the spectral efficiency(SE)and energy efficiency(EE)of the cell-free massive MIMO systems utilizing distributed signal detection on the basis of the analysis of the spatial correlation model of the cell-free massive MIMO systems.It is found that the channel spatial correlation reduces the SE and EE.Then,in view of the low SE of the traditional distributed signal detection,three joint signal detection schemes that utilize channel statistics including large-scale fading decoding(LSFD),local-LSFD(L-LSFD)and optimal bilinear equalizer(OBE)are proposed in this thesis and the SEs,fronthaul load and computational complexity of the three schemes are compared.Compared with the traditional scheme,OBE,LSFD and L-LSFD can improve the SE of the cell-free massive MIMO systems by 1.2 times,1 times and 10% respectively.However,the fronthaul load and computational complexity of OBE are higher compared with LSFD and L-LSFD.Finally,this thesis studies the downlink coherent and non-coherent transmissions of the cell-free massive MIMO systems.Considering the computational complexity of the existing power allocation schemes,the entanglement of the power allocation coefficients is not conducive to the scalability of the cell-free massive MIMO systems.A heuristic power allocation scheme based on channel statistics is present in this thesis.Compared with the traditional power allocation scheme,the heuristic power allocation scheme can improve the SE of the coherent or non-coherent transmission respectively by about 3-6 times.In conclusion,this thesis analyzes the effects of spatial correlation,signal detection and power allocation on the performance of the cell-free massive MIMO systems.The results show that spatial correlation reduces the SE,while the joint signal detection schemes and heuristic power allocation scheme can effectively improve the SE of the cell-free massive MIMO systems.The research results of this thesis can provide theoretical guidance and useful references for the design and deployment of the cell-free massive MIMO systems in practice.