Resource allocation and scheduling in MIMO-OFDMA with time-varying channel and bursty arrivals

A cross-layer resource allocation and scheduling scheme in the downlink of MIMO-OFDMA systems is proposed to provide quality of service (QoS) guarantees to real-time traffics and non-real-time traffics in case of heavy-traffic. The proposed algorithm aims to maximize the efficient system throughput with the constraints of packet error rate, delay, packets dropping rate and buffer size. At the medium access control (MAC) layer, the proposed algorithm dynamically assigns subcarriers to the highest priority user based on the channel state information (CSI) and the queue state information (QSI). After initial subcarrier assignment, the subcarrier re-assignment is applied to approve the system performance. At the physical (PHY) layer, a low complexity rate-and-power adaptation scheme is used to the predetermined subcarriers. It is demonstrated in the numerical results that the proposed algorithm significantly improves the system performance in terms of system throughput and packet loss rate.