Impact of Power Normalization on System-Level Performance in MU-MIMO with User Scheduling

This paper investigates the system-level performance of multi-user MUMO (MU-MEMO) with different power normalization in consideration of user scheduling. En this paper, matrix normalization (MN) and vector normalization (VN) are taken as the power normalization methods. As for user scheduling, semiorthogonal user selection (SUS) or proportional fairness (PF) is adopted, because they are opposed to each other in terms of the quality indexes, such as system capacity and fairness. Specifically, we evaluate the impact of these power normalization methods on the quality indexes and clarify the favorable matching between the power normalization and user scheduling to achieve certain quality indexes.