Robust packet scheduling in wireless cellular networks

This paper addresses the scheduling problem in wireless packet switched networks. Given that only coarse-grained channel state information is available, how do we design a robust scheduler that ensures worst-case optimal performance? To solve this problem, we take a zero-sum game theoretic approach, in which the scheduler and the channel error act as non-cooperative adversaries in the scheduling process. The scheduler tries to maximize the system revenue from serving flows while the channel errors try to minimize it. We give the optimal scheduling strategy in the heavy channel error case and the mild channel error case separately. Such an optimal scheduling strategy is probabilistic, in the sense that the scheduler makes decision based on a probabilistic distribution. Our results show that this probabilistic scheduling may lead to higher worst-case performance compared to traditional deterministic policies, especially when the channel errors are dynamic and difficult to predict.