Statistical Beamforming for Interference Mitigation in Multi-cell Massive MIMO Systems

This paper proposes a statistical beamforming approach to mitigate interference in multi-cell massive multiple-input multiple-output systems. The proposed approach projects the signal subspace onto the null space of the interference to form the weight vector with the statistical channel state information. In contrast to the existing methods which only utilize the partial null space of the interference, the proposed approach can collect more signal when mitigating the interference. In addition, the large system analysis demonstrates that the interference is eliminated with uniform rectangular arrays. Numerical results show that the proposed approach can achieve higher sum rate than the existing methods and verify the performance analysis.