On probabilistic queue length based joint uplink/downlink resource allocation in OFDMA networks

The emergence of state-of-the-art OFDMA-based wireless standards paved the way to new services with bi-directional requirements. In this paper, we present a probabilistic channel-aware, queue-aware and service-aware joint uplink/ downlink resource allocation scheme. The approach consists of coupling the queue behavior probabilistically by minimizing the divergence between the queue distribution at the base station and mobile sides. Then, a CDF-based resource allocation problem is devised taking into account the channel distribution and the derived queue length distribution. Simulation results show notable performance gains in terms of coupling and a good compromise between fairness and rate.