Research On Precoding And Power Allocation Algorithms Of MBO In Millimeter Wave Massive MIMO-NOMA System

The millimeter-wave band has abundant spectrum resources that sub-6G does not have,but suffers from high transmission loss.Beamforming,as a key technology of massive multiple input multiple output(mMIMO)systems,can just make up for this.The wavelength also facilitates the deployment of large-scale antennas.The combination of millimeter-wave and mMIMO technology has become one of the key technologies of 5G.The current mainstream millimeter-wave mMIMO systems use hybrid precoding or beamspace MIMO architectures to reduce the number of radio frequency(RF)chains to reduce system energy consumption and hardware complexity.Different from orthogonal multiple access technology,non-orthogonal multiple access(NOMA)technology uses signal superposition coding in power domain or code domain to provide services for more users in limited time-frequency resources to achieve large-scale connections and low delayed data transmission further improves system spectral efficiency.Therefore,the integration of millimeter-wave mMIMO and NOMA technology,namely millimeter-wave mMIMO-NOMA technology,will provide a higher frequency spectrum for future mobile communications and provide a strong guarantee for the vision of the Internet of Everything.Aiming at the problem of reliable communication caused by the traditional single beam scheme(SBS)that does not consider the limitation of beamwidth,that is,users covered by millimeter-wave sidelobe beams will suffer severe signal attenuation,and sometimes even cannot maintain reliable communication connections.User pairing and hybrid precoding scheme based on multi-beam scheme(MBS)is proposed.First,since MBS allows users to be paired at any angle of departure(AOD),users with a large difference in equivalent channel gain should be paired as much as possible under the premise of considering beam separation.To reduce the complexity and improve the fairness,the antenna is allocated according to the principle of average allocation,and the channel phase information of the user is extracted to design analog precoding.Finally,by solving the digital precoding optimization problem to maximize the spectral efficiency,Since the NOMA users under MBS are no longer similar AODs,the traditional zero-forcing(ZF)design scheme cannot eliminate the inter-group user interference for some users.Therefore,the monarch butterfly optimization(MBO)algorithm is used to solve the non-convex problem.Experiments show that this scheme can achieve better spectral efficiency performance compared with existing schemes while ensuring reliable multi-user communication.Maximizing spectral efficiency as the optimization goal will cause the system to concentrate all resources on users with better channels,and it is difficult to ensure the quality of service for users with poor channels.Therefore,based on the beamspace MIMO-NOMA system,this paper proposes a power allocation scheme based on maximizing the minimum rate.First,considering that the traditional beam selection algorithm cannot be directly applied to the NOMA system,a beam selection method based on the improved maximized channel gain(MM-S)criterion is proposed to reduce the system dimension and realize user grouping.Finally,an optimization problem that maximizes the minimum rate is modeled,and an MBO-based power allocation algorithm is used to solve this non-convex problem.The experimental results show that compared with the existing schemes,the minimum user rate can be effectively improved,thereby ensuring user fairness.