Performance evaluation of layer adaptive multi-user scheduling in LTE-A downlink

Multi-user MIMO is an important technique to utilize the spatial degree of freedom brought by multiple antennas at the transmitter. To be employed in the downlink of practical systems, some sub-optimal algorithms have to be developed to handle the user selection and precoding with limited feedback on channel state information and reasonable computational complexity at a base station. In this paper we evaluate the performance of a specific multi-user scheduling algorithm, which selects users one by one with greedy search and results in a scheduled user set with adaptive size. The performance is compared with traditional two-user scheduling algorithm in single-cell, and we also show the performance of Coordinated Multi-Point transmission with the layer adaptive algorithm by applying it to users from multiple cells. The system level simulation is based on 3GPP LTE-A framework, and the evaluation can be used as guidance for concrete scheduler design.