Weighted Sum-Rate Maximization for Analog Beamforming and Combining in Millimeter Wave Massive MIMO Communications

The analog beamforming and analog combining in millimeter-wave (mm-Wave) massive MIMO communications are investigated. Based on a hierarchical codebook, the weighted sum-rate maximization is studied by jointly considering the duration for channel training and the receiving signal-to-noise ratio (SNR). An algorithm employing exhaustive search at the starting level of the codebook and the multi-sectional search at the other levels is proposed. At each iteration, the channel gain of the dominant path is estimated and the level of the codebook achieving the weighted sum-rate maximization is predicted by the receiver. Then the predicted level is fed back to the transmitter. If the predicted level is reached by the algorithm, the transmitter begins the data transmission with the obtained analog beamformer and the receiver uses the obtained analog combiner for data receiving. Simulation results show that the proposed algorithm outperforms multi-sectional search, where the former can achieve the weighted sum-rate almost twice of the latter when each block is consisted of 35 symbols and the channel SNR is 20 dB.