Design And Analysis Of Scalable Cell-Free Massive MIMO Systems With Hybrid Precision ADC Over Spatially Correlated Rician Fading Channels

With the development of wireless communication technologies,the fifth generation(5G)mobile communication systems have been widely applied in the world.Nowadays,research institutions and companies around the world have also started the research on 6th generation(6G)mobile communication technologies.With the densification of communication networks,the inter-cell interference severely limits the system capacity.Scalable cell-free massive MIMO technology,in which many distributed access points(APs)simultaneously serve UEs by using coherent signal processing method,can provide user-centric uniform coverage and is hence considered as a promising candidate technology for 6G.However,the channel spatial correlation characteristics between different antennas and the line of sight(LOS)path between transceivers are often ignored in many existing works,which makes the analysis results less practical.In addition,despite the technical advantages of scalable cell-free massive MIMO,it still faces two main challenges: high detection complexity and high hardware cost.In this thesis,we consider both the hybrid precision Analog-to-Digital Converter(ADC)and the correlated Rician fading channels in scalable cell-free massive MIMO systems.Firstly,this thesis investigates the performance analysis and the signal detection of scalable cell-free massive MIMO systems with hybrid precision ADC under correlated Rician fading channels.By combining the scalable characteristics of the system,a low-complexity detection scheme for distributed architecture is proposed,which combines the maximal-ratio combining(MRC)and the large-scale fading decoding(LSFD)to design the decoding vector according to the results of AP selection,thus reducing the computational complexity of the decoding vector.The closed-form expressions of the uplink spectral efficiency with imperfect channel state information(CSI)condition are derived.Although the conventional MRC detector has lower computational complexity,it cannot suppress the inter-user interference very well.Thus,low-complexity minimum mean square error(MMSE)detectors are proposed for distributed and centralized schemes to improve the system performance.Secondly,the energy efficiency of the considered system with low precision DAC at the transmitter has been investigated and a joint algorithm consisting of AP cluster formation,pilot assignment,and power control has been studied.Specifically,the power consumption model of the considered system has been established,and the trade-off between the energy efficiency and spectral efficiency has also been investigated.In order to suppress the pilot contamination caused by the UEs sharing the same pilot and simultaneously achieve the desirable tradeoff between the Qo S fairness and average SE,this thesis investigates a joint algorithm consisting of AP selection,pilot assignment and power control for data transmission.Simulation results show that,on one hand,the proposed detectors and weighting vectors are scalable and their spectral efficiency performances are very close to those obtained by equivalent unscalable schemes.The studied pilot assignment scheme yields much higher spectral efficiency performance as compared with the random pilot assignment scheme.On the other hand,the Qo S fairness of each UE can be greatly enhanced as compared to the traditional equal power transmit scheme.