Research On Hybrid Beamforming Technology In 5G Massive MIMO

The millimeter wave in fifth generation(5G)mobile broadband network makes it possible to package massive antennas with small sizes,and the high beamforming gain brought by the massive antenna array greatly alleviates the path depletion of millimeter wave signal transmission,so massive MIMO and beamforming become the key technology in 5G communication.Due to the sheer size of the antennas,the need for a dedicated radio frequency(RF)chain for each antenna in fully digital beamforming would lead to excessive power consumption of the system.Although analog beamforming appears to be a potential alternative,it cannot support multiple streams of signals.For this reason,in this paper,the corresponding hybrid beamforming algorithms are studied and designed for the number of users at the receiver end of hybrid beamforming,and the specific work and contributions of this thesis are as follows:(1)For single-user low-resolution phase-shifter MIMO systems,this paper proposes a low-resolution hybrid beamforming algorithm based on the existing iterative phasematching-based beamforming algorithm using the recursive idea.The proposed algorithm differs from the existing schemes by randomly initializing the analog beamformer,where each analog precoder and combiner vector is reasonably optimized to solve the problem of local optimum caused by conventional algorithms.At the same time,our scheme does not require internal and external iterative processes compared to existing schemes,thus effectively reducing the complexity of the algorithm.Simulation results show that the proposed algorithm improves the performance while reducing the complexity to onefortieth of the comparison scheme compared to the iterative phase matching algorithm.In addition,considering the complexity imposed by full connectivity and imperfect CSI scenarios,it is observed that both analog and digital precoders exhibit near orthogonality,so that a low-complexity dot matrix decoder based on the Babai algorithm can be used.To further improve the achievable spectral efficiency as well as the energy efficiency,we randomly deactivate some phase shifters,similar to the approach used in spatial modulation,which incorporates a disconnected point 0 in the designed constellation.Based on this principle,the feasible domain in the finite-resolution phase shifter is divided into a sub-circular set and a parallel set of zeros.Simulation results likewise demonstrate that this structure can effectively improve the spectral efficiency.(2)For the multi-user low-resolution phase shifter MIMO system,instead of randomly initialzing the analog precoding matrix(?) and analog combining matrix(?),we initialize them according to the beam arrival and beam departure angles.After the initialization,the coarse search beam set is constructed.That is,the array response vector is quantized to meet the low-resolution of the phase shifter.After obtaining the coarse search beam set at the transceiver,a joint search is performed to find the coarse beam matrix with the maximum order and rate,and finally the optimal beamforming matrix is found using the local search algorithm.The simulation results show that the algorithm reduces the complexity by half and has a better sum rate than the existing schemes.