Research On High-performance Beamforming In Massive MU-MIMO

Massive multiple-input multiple-output(MIMO)with multi-users is widely recognized as an promising means to enhance system capacity for the fifth generation of wireless communications.A large number of spatial degrees of freedom acquired by large-scale antenna arrays at base stations,could be exploited to improve system performance.However,the problem of inter-cell interference in massive MU-MIMO systems is more and more serious,which has become one of the primary reasons to restrain the performance of the system.In this paper,we propose an algorithm named dynamic division of the areas cooperative beamforming(DDA-CoBF)to suppress the interference of cell edge users.In addition,we propose two kinds of division schemes of collaborative areas,square and circular,based on DDA-CoBF algorithm to divide cooperative users out of the cell and to conclude the optimal DDA-CoBF scheme of collaborative region division for the energy efficiency.Comparing with other CoMP methods,the base applied with DDA-CoBF algorithm not only guarantee the efficient coverage of users,but also improve the energy efficiency and throughput.In the end,considering the difficulties in obtaining channel state information and the shortage of pilot frequency range in massive MU-MIMO system,we analyse the influence of CSI error and pilot pollution on DDA-CoBF algorithm.The results show that DDA-CoBF algorithm is susceptible to estimation error interference(CSI).Therefore this paper proposes an estimation error prediction(EEP)algorithm to optimizes the precoding vector by reducing the mean square error(MSE)between the transmitter and receiver.The simulation results show that the strong robustness DDA-CoBF algorithm could effectively reduce the bit error rate of the users,especially improve the performance at low SINR.