Efficient periodic scheduling by trees

In a perfectly-periodic schedule, time is divided into time-slots, and each client gets a time slot precisely every predefined number of time slots. The input to a schedule design algorithm is a frequency request for each client, and its task is to construct a perfectly periodic schedule that matches the requests as "closely" as possible. The quality of the schedule is measured by the ratios between the requested frequency and the allocated frequency for each client (either by the weighted average or by the maximum of these ratios over all clients). Periodic schedules enjoy maximal fairness, and are very useful in many contexts of asymmetric communication, e.g., push systems and Bluetooth networks. However, finding an optimal periodic schedule is NP-hard in general. Tree scheduling is a methodology for developing perfectly periodic schedules with quality guarantees by constructing trees that correspond to periodic schedules. We explore a few aspects of tree scheduling. First, noting that a complete schedule table may be exponential in size, and that using the tree for scheduling directly may require logarithmic time on average, we give algorithms that find the next client to schedule in constant amortized time, using only polynomial space in most practical cases. Second, we present a few heuristic algorithms for generating schedules, based on analysis of optimal tree-scheduling algorithms, for both the average and maximum measures. Simulation results indicate that some of these heuristics produce excellent schedules in practice, sometimes even beating the best known non-periodic schedules.