Multiuser Scheduling to Improve Iterative Interference Replica Subtraction with Partial Zeroization for Massive MIMO Systems

In massive MIMO, precoding has an interference cancellation issue because the channel size has a considerable dimension based on the number of antennas and receivers. This paper proposes applying a simple user scheduling to improve the signal-to-interference-plus-noise power ratio (SINR) of the previously proposed lightweight precoding scheme. As the complexity-efficient precoding for single-cell multiuser massive MIMO systems, an iterative interference replica signal subtraction (IIRS) was investigated. To further reduce computational complexity and to save the worst-case performance of IIRS, we previously suggested zeroizing the minute replica component. This approach prevents unnecessary interference cancellation that could be affected by errors in channel estimation. Multiuser scheduling is based on the spatial correlation among users, which is suitable for IIRS because the correlation corresponds to replica components. Simulative evaluation verifies its effectiveness; user scheduling benefits SINR and computation reduction.