Cross-layer Scheduling and Dynamic Resource Allocation for Downlink Multiuser MIMO Systems with Limited Feedback

In this paper, we present a cross-layer scheduling and dynamic resource allocation (CLSDRA) strategy for downlink multiuser multiple-input multiple-output (MIMO) systems with limited feedback. Based on the instantaneous partial channel state information (CSI) from physical (PHY) layer and the queue state information (QSI) from media access control (MAC) layer, the proposed scheme schedules packets and allocates resources such as channel, bit and power to the scheduled users to maximize the utility function which sums the product of each scheduled user's delay requirement, head of line (HOL) packet waiting time, queue length and data rate. Simulation results demonstrate that our scheme outperforms the existing single-layer resource allocation or scheduling algorithms in balancing the total system throughput and the packet drop rate because the proposed strategy could utilize the QSI and the partial CSI simultaneously.