Multi-cell massive MIMO coordinated beamforming for transmit power minimization with imperfect CSIT

In this paper, we propose a scheme to design transmit beamforming considering the effect of channel estimation at the transmitter. Under the multi-cell downlink massive MIMO system, coordinated beamforming is obtained by applying uplink-downlink duality for transmit power minimization problem subject to signal-to-interference-plus-noise ratio (SINR) constraints. To handle much information exchange and cumbersome computational complexity for obtaining coordinated beamforming, we derive asymptotic equivalent (AE) of SINR for uplink and downlink system by employing random matrix theory. The AE depends on the statistical channel state information (CSI) and intra/inter-cell channel estimation error. According to asymptotic results, our proposed scheme can efficiently reduce signaling overhead with less power information and CSI exchange. Simulation results show that the proposed method performs well in transmit power and possesses fast convergence rate.