Adaptive SU/MU-MIMO scheduling schemes for LTE-A downlink transmission

We investigate multi-user (MU) multiple-input multiple-output (MIMO) scheduling under the practical constraints in long-term evolution advanced (LTE-A) downlink cellular networks. The authors first derive the received signal model in MU-MIMO systems when there exist both inter-stream interference (ISI) and inter-user interference (IUI) at each user device. Based on this, they formulate the optimisation problem as joint user pairing, precoding matrix indicator (PMI) selection, and resource block (RB) allocation to maximise the total system throughput. The authors then develop a codebook grouping technique for user pairing and PMI selection in MU-MIMO. With the help of codebook grouping, the authors propose some suboptimal and low-complexity scheduling algorithms to improve system throughput. From system-level simulation, the proposed algorithms can improve the network throughput significantly when exploiting only limited feedback designed for single-user (SU) MIMO in the LTE-A specification.