FlexSched: A Parameterized Data Schedule Generator for Multi-Channel Broadcast Systems

In this paper we study the problem of finding an efficient schedule for broadcasting a given set of data items over multiple physical broadcast channels. The goal of our work is twofold: (a) to minimize the average access latency to fulfill data requests of mobile clients and (b) to provide the ability to tune the period length of the broadcast schedule so as to ease the air-indexing process and to enable clients to determine when the desired data item is broadcast next. To achieve these goals, we propose a parameterized scheduler, called FlexSched, which is a polynomial time algorithm and uses a divide-and-conquer approach to efficiently obtain a solution to the problem. To evaluate the robustness and solution quality of FlexSched, we conducted an extensive performance study by using several synthetic data sets. The results show that FlexSched may significantly outperform existing state-of-the-art data scheduling algorithms with relatively low computational costs.