Performance of greedy policies for downlink scheduling in networks with relay stations

Abstract

We investigate a greedy policy for delay-optimal scheduling in broadband wireless networks using relay stations (RS). The key idea behind the policy is that users should greedily exploit opportunism from time varying channels, while the basestation should prioritize balancing queues at RS. A fluid model is presented to show that such policy is optimal in the fluid regime. By simulation we show that the proposed scheme reduces mean delay by up to 73% compared to backpressure algorithm.