ABSTRACT:

In this paper, we tend to are fascinated by wireless scheduling algorithms for the downlink of one cell which will minimize the queue-overflow probability. Specifically, in a very giant-deviation setting, we are fascinated by algorithms that maximize the asymptotic decay rate of the queue-overflow likelihood, as the queue-overflow threshold approaches infinity. We tend to initial derive an upper bound on the decay rate of the queue-overflow likelihood over all scheduling policies. We have a tendency to then focus on a class of scheduling algorithms collectively known as the “ -algorithms.” For a given the -algorithm picks the user for service at each time that has the largest product of the transmission rate multiplied by the backlog raised to the ability. We have a tendency to show that when the overflow metric is appropriately modified, the minimum-cost-to-overflow below the -algorithm will be achieved by a straightforward linear path, and it can be written as the answer of a vector-optimization downside. Using this structural property, we have a tendency to then show that when approaches infinity, the -algorithms asymptotically achieve the most important decay rate of the queue-overflow probability. Finally, this result enables us to design scheduling algorithms that are each shut to optimal in terms of the asymptotic decay rate of the overflow likelihood and empirically shown to take care of tiny queue-overflow chances over queue-length ranges of practical interest.