Multiple-Beam Selection with Limited Feedback for Hybrid Beamforming in Massive MIMO Systems
Yuwei Ren,Yingmin Wang,Can Qi,Yinjun Liu
DOI: https://doi.org/10.1109/access.2017.2666782
IF: 3.9
2017-01-01
IEEE Access
Abstract:Hybrid multiple-input multiple-output (MIMO) systems have been thought as a promising technology in future 5G. Compared with conventional digital MIMO systems, such a structure is equipped with fewer RF chains, which would reduce the computational complexity and hardware cost, and meanwhile additional analog beamforming is introduced to maintain the performance. However, scarce RF chains make channel state information acquisition difficult for analog beamforming. In this paper, we consider a practical hybrid beamforming, which includes zero-forcing (ZF) precoding in digital beamforming, and beam selection for analog beamforming. First, the statistic information of users (e.g., angle and distance) is utilized to construct an approximate channel for each user. Second, users individually evaluate the codebook and feedback the results, by which base station (BS) makes optimization and selects the final beams for analog beamforming. Finally, BS performs the digital baseband ZF precoding with the equivalent channel. In the process, we give two limited feedback methods for users, and two corresponding beam selection methods for BS. These methods are evaluated in the Rayleigh fading channels and mmWave channels. Simulation results show that our hybrid beamforming could approach performance of conventional digital precoding, and more RF chains could provide better performance. Moreover, proposed methods only require once feedback and effectively reduce the delay, and two feedback methods achieve a good tradeoff between performance and feedback cost.