Prioritized Maximal Scheduling in Wireless Networks

This paper considers the scheduling problem in wireless networks. We focus on prioritized maximal scheduling, where a maximal scheduler chooses the links in an order specified by certain priority. We first analyze the capacity region of a maximal scheduler with fixed priority, where a lower bound is formulated and shown to be tight. Next we propose both centralized and distributed algorithms to search for the optimal priority for a given arrival rate. The algorithms are optimal in the sense that, if the arrival rate is in the lower bound region of any prioritized maximal scheduling, it is stable under the result of our algorithms. Finally, by combining the priority assignment and maximal scheduling, we prove that we can achieve a certain fraction of the optimal capacity region, which is bounded below by a constant for most networks.