Research On NOMA-Based Cell-Free Massive MIMO Systems

With the development of the research on the 6th Generation(6G)mobile communication system,cell-free multi-input multiple-output(MIMO)technology has attracted wide attention as a candidate technology.It can overcome the problems of poor performance of the users near the cell edge,frequent handover,and serious interference in the cellular system.Moreover,non-orthogonal multiple access(NOMA)is a key technology in the future development of wireless communication,which can meet the needs of high spectral and energy efficiency,low latency,massive access,and user fairness in the next generation of mobile communication.Therefore,the combination of the two technologies is expected to reduce interference while guaranteeing massive access,and then,effectively improve the spectral efficiency(SE)and energy efficiency(EE)of the system.Based on this,the main research contents of this paper can be summarized as follows:Firstly,this paper considers a NOMA-based cell-free massive MIMO system over spatially correlated Ricain fading channels,and the theoretical framework,signal transmission process,and user decoding order for uplink performance analysis are proposed.Applying minimum mean-square error(MMSE),element-wise MMSE(EW-MMSE),least-square(LS)channel estimations and according to the channel statistical characteristics,the closed-form expressions of the uplink SE are derived,respectively.The effects of MMSE,EW-MMSE,and LS estimations,channel parameters,access points(APs)configuration,and deployment mode on the system performance are analyzed by simulation and analytical results.It demonstrates that NOMA can achieve approximately 50% improvement of the sum SE compared to OMA in cell-free massive MIMO system.Secondly,because of the differences in the transmitting power,the user decoding order,the interference received by users and the difficulty of implementation between the uplink and downlink NOMA,this paper further analyzes the system downlink performance.According to the information theory and large dimensional random matrices theory,the closed-form expressions of downlink SE and EE are derived and the validity of the analytical results are proved through the simulation results.This paper analyses the effects of the channel estimations,channel fading models,imperfect successive interference cancelation,AP antenna configuration,channel correlation,and precoding methods on the downlink SE.The results show that when the number of users is greater than 90,the performance of NOMA scheme will outperform OMA with 40 APs.Finally,in order to further improve the system performance of NOMA-based cell-free massive MIMO,the system optimization in this paper is designed from three aspects: user pairing based on large-scale information,uplink optimal bilinear equalizer design,and downlink power allocation optimization based on convex optimization theory.The simulation results demonstrate the feasibility of user pairing,uplink optimal bilinear equalizer design,and downlink power allocation optimization algorithm in improving the system performance.The average SE is improved by about 35% after using the power allocation algorithm.In conclusion,this paper studies the NOMA-based cell-free MIMO system.It shows the closed-form derivations of the key performance indicators such as uplink and downlink SE and EE,and reveals the effects of various parameters and signal processing methods on the system performance from a multi-dimensional perspective.Then,the system is optimized from three aspects: user pairing,uplink joint signal detection,and downlink power allocation.The research provides a theoretical basis for the design of a NOMA-based cell-free massive MIMO system and contributes to the development of 6G mobile communication system in the future.